Amazon Bedrock input and output pricesUSD per 1M tokens · log scale
$0.01$0.10$1.00$10.00$100.00Nova MicroNova Micro · Input: $0.035Nova Micro · Output: $0.14Gemma 4 E2BGemma 4 E2B · Input: $0.04Gemma 4 E2B · Output: $0.08Nemotron Nano 3 30BNemotron Nano 3 30B · Input: $0.06Nemotron Nano 3 30B · Output: $0.24Nova LiteNova Lite · Input: $0.06Nova Lite · Output: $0.24GPT OSS Safeguard 20BGPT OSS Safeguard 20B · Input: $0.07GPT OSS Safeguard 20B · Output: $0.20gpt-oss-20bgpt-oss-20b · Input: $0.07gpt-oss-20b · Output: $0.30Ministral 3B 3.0Ministral 3B 3.0 · Input: $0.10Ministral 3B 3.0 · Output: $0.10Gemma 4 26B A4BGemma 4 26B A4B · Input: $0.13Gemma 4 26B A4B · Output: $0.40Gemma 4 31BGemma 4 31B · Input: $0.14Gemma 4 31B · Output: $0.40Qwen3 Next 80B A3BQwen3 Next 80B A3B · Input: $0.14Qwen3 Next 80B A3B · Output: $1.20GPT OSS Safeguard 120BGPT OSS Safeguard 120B · Input: $0.15GPT OSS Safeguard 120B · Output: $0.60gpt-oss-120bgpt-oss-120b · Input: $0.15gpt-oss-120b · Output: $0.60Ministral 8B 3.0Ministral 8B 3.0 · Input: $0.15Ministral 8B 3.0 · Output: $0.15NVIDIA Nemotron 3 Super 120B A12BNVIDIA Nemotron 3 Super 120B A12B · Input: $0.15NVIDIA Nemotron 3 Super 120B A12B · Output: $0.65Qwen3 32BQwen3 32B · Input: $0.15Qwen3 32B · Output: $0.60Writer Palmyra Vision 7BWriter Palmyra Vision 7B · Input: $0.15Writer Palmyra Vision 7B · Output: $0.60Ministral 14B 3.0Ministral 14B 3.0 · Input: $0.20Ministral 14B 3.0 · Output: $0.20Qwen3 235B A22B 2507Qwen3 235B A22B 2507 · Input: $0.22Qwen3 235B A22B 2507 · Output: $0.88MiniMax M2.5MiniMax M2.5 · Input: $0.30MiniMax M2.5 · Output: $1.20Nova 2.0 LiteNova 2.0 Lite · Input: $0.33Nova 2.0 Lite · Output: $2.75Magistral Small 1.2Magistral Small 1.2 · Input: $0.50Magistral Small 1.2 · Output: $1.50Mistral Large 3Mistral Large 3 · Input: $0.50Mistral Large 3 · Output: $1.50Qwen3 Coder NextQwen3 Coder Next · Input: $0.50Qwen3 Coder Next · Output: $1.20Qwen3 VL 235B A22BQwen3 VL 235B A22B · Input: $0.53Qwen3 VL 235B A22B · Output: $2.66Kimi K2.5Kimi K2.5 · Input: $0.60Kimi K2.5 · Output: $3.00DeepSeek V3.2DeepSeek V3.2 · Input: $0.62DeepSeek V3.2 · Output: $1.85Nova ProNova Pro · Input: $0.80Nova Pro · Output: $3.20GLM 5GLM 5 · Input: $1.00GLM 5 · Output: $3.20Mistral SmallMistral Small · Input: $1.00Mistral Small · Output: $3.00Grok 4.3Grok 4.3 · Input: $1.25Grok 4.3 · Output: $2.50Pixtral Large 25.02Pixtral Large 25.02 · Input: $2.00Pixtral Large 25.02 · Output: $6.00Grok 4.6Grok 4.6 · Input: $2.20Grok 4.6 · Output: $6.60Grok 4.7Grok 4.7 · Input: $2.20Grok 4.7 · Output: $6.60Nova PremierNova Premier · Input: $2.50Nova Premier · Output: $12.50Kimi K3Kimi K3 · Input: $3.30Kimi K3 · Output: $16.50Mistral LargeMistral Large · Input: $4.00Mistral Large · Output: $12.00$0.01$0.10$1.00$10.00$100.00Nova MicroNova Micro · Input: $0.035Nova Micro · Output: $0.14Gemma 4 E2BGemma 4 E2B · Input: $0.04Gemma 4 E2B · Output: $0.08Nemotron Nano 3 30BNemotron Nano 3 30B · Input: $0.06Nemotron Nano 3 30B · Output: $0.24Nova LiteNova Lite · Input: $0.06Nova Lite · Output: $0.24GPT OSS Safeguard 20BGPT OSS Safeguard 20B · Input: $0.07GPT OSS Safeguard 20B · Output: $0.20gpt-oss-20bgpt-oss-20b · Input: $0.07gpt-oss-20b · Output: $0.30Ministral 3B 3.0Ministral 3B 3.0 · Input: $0.10Ministral 3B 3.0 · Output: $0.10Gemma 4 26B A4BGemma 4 26B A4B · Input: $0.13Gemma 4 26B A4B · Output: $0.40Gemma 4 31BGemma 4 31B · Input: $0.14Gemma 4 31B · Output: $0.40Qwen3 Next 80B A3BQwen3 Next 80B A3B · Input: $0.14Qwen3 Next 80B A3B · Output: $1.20GPT OSS Safeguard 120BGPT OSS Safeguard 120B · Input: $0.15GPT OSS Safeguard 120B · Output: $0.60gpt-oss-120bgpt-oss-120b · Input: $0.15gpt-oss-120b · Output: $0.60Ministral 8B 3.0Ministral 8B 3.0 · Input: $0.15Ministral 8B 3.0 · Output: $0.15NVIDIA Nemotron 3 Super 120B A12BNVIDIA Nemotron 3 Super 120B A12B · Input: $0.15NVIDIA Nemotron 3 Super 120B A12B · Output: $0.65Qwen3 32BQwen3 32B · Input: $0.15Qwen3 32B · Output: $0.60Writer Palmyra Vision 7BWriter Palmyra Vision 7B · Input: $0.15Writer Palmyra Vision 7B · Output: $0.60Ministral 14B 3.0Ministral 14B 3.0 · Input: $0.20Ministral 14B 3.0 · Output: $0.20Qwen3 235B A22B 2507Qwen3 235B A22B 2507 · Input: $0.22Qwen3 235B A22B 2507 · Output: $0.88MiniMax M2.5MiniMax M2.5 · Input: $0.30MiniMax M2.5 · Output: $1.20Nova 2.0 LiteNova 2.0 Lite · Input: $0.33Nova 2.0 Lite · Output: $2.75Magistral Small 1.2Magistral Small 1.2 · Input: $0.50Magistral Small 1.2 · Output: $1.50Mistral Large 3Mistral Large 3 · Input: $0.50Mistral Large 3 · Output: $1.50Qwen3 Coder NextQwen3 Coder Next · Input: $0.50Qwen3 Coder Next · Output: $1.20Qwen3 VL 235B A22BQwen3 VL 235B A22B · Input: $0.53Qwen3 VL 235B A22B · Output: $2.66Kimi K2.5Kimi K2.5 · Input: $0.60Kimi K2.5 · Output: $3.00DeepSeek V3.2DeepSeek V3.2 · Input: $0.62DeepSeek V3.2 · Output: $1.85Nova ProNova Pro · Input: $0.80Nova Pro · Output: $3.20GLM 5GLM 5 · Input: $1.00GLM 5 · Output: $3.20Mistral SmallMistral Small · Input: $1.00Mistral Small · Output: $3.00Grok 4.3Grok 4.3 · Input: $1.25Grok 4.3 · Output: $2.50Pixtral Large 25.02Pixtral Large 25.02 · Input: $2.00Pixtral Large 25.02 · Output: $6.00Grok 4.6Grok 4.6 · Input: $2.20Grok 4.6 · Output: $6.60Grok 4.7Grok 4.7 · Input: $2.20Grok 4.7 · Output: $6.60Nova PremierNova Premier · Input: $2.50Nova Premier · Output: $12.50Kimi K3Kimi K3 · Input: $3.30Kimi K3 · Output: $16.50Mistral LargeMistral Large · Input: $4.00Mistral Large · Output: $12.00
Amazon Bedrock input and output prices
CategoryItemUSD per 1M tokens
Nova MicroInput$0.035
Nova MicroOutput$0.14
Gemma 4 E2BInput$0.04
Gemma 4 E2BOutput$0.08
Nemotron Nano 3 30BInput$0.06
Nemotron Nano 3 30BOutput$0.24
Nova LiteInput$0.06
Nova LiteOutput$0.24
GPT OSS Safeguard 20BInput$0.07
GPT OSS Safeguard 20BOutput$0.20
gpt-oss-20bInput$0.07
gpt-oss-20bOutput$0.30
Ministral 3B 3.0Input$0.10
Ministral 3B 3.0Output$0.10
Gemma 4 26B A4BInput$0.13
Gemma 4 26B A4BOutput$0.40
Gemma 4 31BInput$0.14
Gemma 4 31BOutput$0.40
Qwen3 Next 80B A3BInput$0.14
Qwen3 Next 80B A3BOutput$1.20
GPT OSS Safeguard 120BInput$0.15
GPT OSS Safeguard 120BOutput$0.60
gpt-oss-120bInput$0.15
gpt-oss-120bOutput$0.60
Ministral 8B 3.0Input$0.15
Ministral 8B 3.0Output$0.15
NVIDIA Nemotron 3 Super 120B A12BInput$0.15
NVIDIA Nemotron 3 Super 120B A12BOutput$0.65
Qwen3 32BInput$0.15
Qwen3 32BOutput$0.60
Writer Palmyra Vision 7BInput$0.15
Writer Palmyra Vision 7BOutput$0.60
Ministral 14B 3.0Input$0.20
Ministral 14B 3.0Output$0.20
Qwen3 235B A22B 2507Input$0.22
Qwen3 235B A22B 2507Output$0.88
MiniMax M2.5Input$0.30
MiniMax M2.5Output$1.20
Nova 2.0 LiteInput$0.33
Nova 2.0 LiteOutput$2.75
Magistral Small 1.2Input$0.50
Magistral Small 1.2Output$1.50
Mistral Large 3Input$0.50
Mistral Large 3Output$1.50
Qwen3 Coder NextInput$0.50
Qwen3 Coder NextOutput$1.20
Qwen3 VL 235B A22BInput$0.53
Qwen3 VL 235B A22BOutput$2.66
Kimi K2.5Input$0.60
Kimi K2.5Output$3.00
DeepSeek V3.2Input$0.62
DeepSeek V3.2Output$1.85
Nova ProInput$0.80
Nova ProOutput$3.20
GLM 5Input$1.00
GLM 5Output$3.20
Mistral SmallInput$1.00
Mistral SmallOutput$3.00
Grok 4.3Input$1.25
Grok 4.3Output$2.50
Pixtral Large 25.02Input$2.00
Pixtral Large 25.02Output$6.00
Grok 4.6Input$2.20
Grok 4.6Output$6.60
Grok 4.7Input$2.20
Grok 4.7Output$6.60
Nova PremierInput$2.50
Nova PremierOutput$12.50
Kimi K3Input$3.30
Kimi K3Output$16.50
Mistral LargeInput$4.00
Mistral LargeOutput$12.00

Prices per 1M tokens. A dash means no single price is listed, or, under capability, no independent score.

API prices by model
ModelInputCache readOutputCapability (Epoch AI)
DeepSeek R1Deprecated$1.35No single price$5.40No independent capability score
DeepSeek V3.1Deprecated$0.58No single price$1.68No independent capability score
DeepSeek V3.2$0.62No single price$1.85No independent capability score
Devstral 2 123BDeprecated$0.40No single price$2.00No independent capability score
Gemma 3 12BDeprecated$0.09No single price$0.29No independent capability score
Gemma 3 27BDeprecated$0.23No single price$0.38No independent capability score
Gemma 3 4BDeprecated$0.04No single price$0.08No independent capability score
Gemma 4 26B A4B$0.13No single price$0.40No independent capability score
Gemma 4 31B$0.14No single price$0.40No independent capability score
Gemma 4 E2B$0.04No single price$0.08No independent capability score
GLM 4.7Deprecated$0.60No single price$2.20No independent capability score
GLM 4.7 FlashDeprecated$0.07No single price$0.40No independent capability score
GLM 5Compare GLM-5 across 3 providers$1.00No single price$3.20Capable · 146.5
GPT OSS Safeguard 120B$0.15No single price$0.60No independent capability score
GPT OSS Safeguard 20B$0.07No single price$0.20No independent capability score
gpt-oss-120bCompare gpt-oss-120b across 6 providers$0.15No single price$0.60Capable · 140.2
gpt-oss-20bCompare gpt-oss-20b across 3 providers$0.07No single price$0.30Entry · 134.9
Grok 4.3$1.25$0.20$2.50No independent capability score
Grok 4.6$2.20$0.55$6.60No independent capability score
Grok 4.7Compare Grok 4.7 across 2 providers$2.20$0.55$6.60Strong · 156.4
Kimi K2 ThinkingDeprecated$0.60No single price$2.50No independent capability score
Kimi K2.5Compare Kimi K2.5 across 4 providers$0.60No single price$3.00Capable · 148.8
Kimi K3Compare Kimi K3 across 7 providers$3.30$0.33$16.50Strong · 158.6
Llama 3 70BDeprecated$2.65No single price$3.50No independent capability score
Llama 3 8BDeprecated$0.30No single price$0.60No independent capability score
Llama 3.1 70BDeprecated$0.72No single price$0.72No independent capability score
Llama 3.1 8BDeprecated$0.22No single price$0.22No independent capability score
Llama 3.3 70BDeprecated$0.72No single price$0.72No independent capability score
Llama 4 Maverick 17BDeprecated$0.24No single price$0.97No independent capability score
Llama 4 Scout 17BDeprecated$0.17No single price$0.66No independent capability score
Magistral Small 1.2$0.50No single price$1.50No independent capability score
MiniMax M2Deprecated$0.30No single price$1.20No independent capability score
MiniMax M2.1Deprecated$0.30No single price$1.20No independent capability score
MiniMax M2.5$0.30No single price$1.20No independent capability score
Ministral 14B 3.0$0.20No single price$0.20No independent capability score
Ministral 3B 3.0$0.10No single price$0.10No independent capability score
Ministral 8B 3.0$0.15No single price$0.15No independent capability score
Mistral 7BDeprecated$0.15No single price$0.20No independent capability score
Mistral Large$4.00No single price$12.00No independent capability score
Mistral Large 3$0.50No single price$1.50No independent capability score
Mistral Small$1.00No single price$3.00No independent capability score
Mixtral 8x7BDeprecated$0.45No single price$0.70No independent capability score
Nemotron Nano 3 30B$0.06No single price$0.24No independent capability score
Nova 2.0 Lite$0.33$0.0825$2.75No independent capability score
Nova Lite$0.06$0.015$0.24No independent capability score
Nova Micro$0.035$0.00875$0.14No independent capability score
Nova Premier$2.50$0.625$12.50No independent capability score
Nova Pro$0.80$0.20$3.20No independent capability score
NVIDIA Nemotron 3 Super 120B A12B$0.15No single price$0.65No independent capability score
NVIDIA Nemotron Nano 2Deprecated$0.06No single price$0.23No independent capability score
NVIDIA Nemotron Nano 2 VLDeprecated$0.20No single price$0.60No independent capability score
Pixtral Large 25.02$2.00No single price$6.00No independent capability score
Qwen3 235B A22B 2507$0.22No single price$0.88No independent capability score
Qwen3 32B$0.15No single price$0.60No independent capability score
Qwen3 Coder 30B A3BDeprecated$0.15No single price$0.60No independent capability score
Qwen3 Coder 480B A35BDeprecated$0.45No single price$1.80No independent capability score
Qwen3 Coder Next$0.50No single price$1.20No independent capability score
Qwen3 Next 80B A3B$0.14No single price$1.20No independent capability score
Qwen3 VL 235B A22B$0.53No single price$2.66No independent capability score
Voxtral Mini 1.0Deprecated$0.04No single price$0.04No independent capability score
Voxtral Small 1.0Deprecated$0.10No single price$0.30No independent capability score
Writer Palmyra Vision 7B$0.15No single price$0.60No independent capability score

Rates, billing conditions and sources

DeepSeek R1

Deprecated by provider · still listed · price checked Nov 12, 2026 (deprecation notice ↗)

All rates, conditions and sources

Conditions of every rate below:

  • Listed for model "R1" (provider "DeepSeek") in us-east-1 with no service_tier attribute (feature "On-demand Inference"), as one USD price dimension from 0 to Inf.
  • AWS lists this price in USD per 1K tokens (unit "1K tokens"); it is multiplied by 1,000 exactly, by decimal arithmetic, to give USD per 1M tokens.
  • AWS's on-demand price for this model with no service tier stated: the row carries no service_tier attribute, so it is not labelled as the standard or any other tier. Only this plain on-demand usagetype in us-east-1 is published: any flex, priority, batch, latency-optimized, custom-model or cross-region (global) variant, other AWS Regions, provisioned throughput and model customization are excluded. The offer file lists no cache price for this model without a service tier; missing cache prices are unknown, not zero.

Input: $1.35 per 1M tokens · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-DeepSeek-R1-input-tokens.
  • Input tokens.

Output: $5.40 per 1M tokens · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-DeepSeek-R1-output-tokens.
  • Output tokens.

Price observed:

API availability: listed as available when checked (not a live service check)

Collection record

Primary source ↗ · provider documentation · Claims: bedrock-deepseek-r1: availability · Retrieved · Checked · Reviewed · Review due · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card for DeepSeek-R1, Model ID deepseek.r1-v1:0, raw SHA-256 ccdc3e7ddd8306fd1526959b475c9e6fc3c61e60d25a60e814d992fa6bfcbff7, read once at 2026-11-02T11:40:04Z under the #837 AWS facts-plus-link basis: "Model lifecycle: Legacy"; "Model EOL date: March 30, 2027"; "Model lifecycle policy: Model lifecycle (For Models Launched Prior to Sept 7 2026)".
  • Amazon Bedrock "Model lifecycle (Legacy)" page (https://docs.aws.amazon.com/bedrock/latest/userguide/model-lifecycle-legacy.html), read once at 2026-11-02T11:37:59Z, raw SHA-256 c33d9d25d7da7f47c098aa864830361de770224f38a790070bada021dac56850: its Legacy table lists DeepSeek-R1 (deepseek.r1-v1:0) with Legacy date September 30, 2026, EOL date March 30, 2027 and public extended access start date "To be announced". The page states: "New customers can't use Legacy models and existing customers may lose access to Legacy models after 15 days of inactivity."
  • That table row's Regions (us-east-1, us-east-2, us-west-2) include us-east-1, the region this offer is published for.
  • Availability stays available: the us-east-1 price list still lists an on-demand price (2026-11-06 archive), and the page says "you can continue to use a Legacy model during this period" (the Legacy period before the EOL date), though not as a new customer. The EOL date is in the future, so no retirement is recorded; reviewAfter is the EOL date, a re-review date, not a retirement.
  • Reviewed by Claude (owner delegation 2026-09-27), #863.

Raw API evidence ↗ · provider API · Claims: bedrock-deepseek-r1: pricing · bedrock-deepseek-r1: availability · 122 more claims about 61 other catalog entries · Retrieved · Checked · Review due · Confidence: A

  • Raw SHA-256: f1ac6a175fb27b3ad80d2c5d8970be8c2e5f957906349e52d15ec11d43e97896; collector aws-bedrock-price-list-json-1.
  • AWS Price List Bulk API offer file for Amazon Bedrock (us-east-1): offer version 20261111013037, publicationDate 2026-11-11T01:30:37Z.
  • Reviewed on-demand token prices only: each curated model's exact usagetypes in us-east-1, with service_tier "standard" for standard-tier models and with no service_tier attribute for the reviewed untiered models, in US dollars as the offer file states them (hence confidence A). A hosted listing and pricing observation only: no checkpoint, developer, license, context-window, modality or release-date assertions, and the usagetype is not a documented Bedrock model ID.
  • No Anthropic Claude price is published from this offer file: prices are read only from curated usagetypes, and no curated entry is an Anthropic model. (When the collector was reviewed on 2026-10-29, the file listed only legacy Claude models, all without a service tier.)
  • Availability rests on AWS's own offer file listing an on-demand price for the model in us-east-1.

DeepSeek V3.1

Context window: 128K tokens

Deprecated by provider · still listed · price checked Nov 12, 2026 (deprecation notice ↗)

All rates, conditions and sources

Conditions of every rate below:

  • Listed for model "DeepSeek V3.1" (provider "DeepSeek") in us-east-1 with service_tier "standard", as one USD price dimension from 0 to Inf.
  • AWS lists this price in USD per 1K tokens (unit "1K tokens"); it is multiplied by 1,000 exactly, by decimal arithmetic, to give USD per 1M tokens.
  • Only the standard on-demand tier in us-east-1 is published: flex, priority, batch and global cross-region tiers, other AWS Regions and provisioned throughput are excluded. The offer file lists no standard-tier cache price for this model; missing cache prices are unknown, not zero.

Input: $0.58 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-deepseek.v3.1-mantle-input-tokens-standard.
  • Input tokens.

Output: $1.68 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-deepseek.v3.1-mantle-output-tokens-standard.
  • Output tokens.

Price observed:

API availability: listed as available when checked (not a live service check)

Collection record

Primary source ↗ · provider documentation · Claims: bedrock-deepseek-v3-1: specifications · Retrieved · Checked · Reviewed · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card ("Models at a glance") for DeepSeek-V3.1, Model ID deepseek.v3.1 (the offer's price list usagetype stem), raw SHA-256 67c72d5b04fcd77f2acd03d2bb1b7c860432073c5d32e65f8ba852d7ea5863a7, read once on 2026-10-31 under the #837 AWS facts-plus-link basis: "Context window: 128K tokens".
  • Shorthand recorded as stated (#504); not converted to an exact token count. The value is AWS's statement for its Bedrock listing; another seller of the same model may state a different figure.
  • Specifications only: no pricing, availability, capability, lifecycle or release-date assertion.

Primary source ↗ · provider documentation · Claims: bedrock-deepseek-v3-1: availability · Retrieved · Checked · Reviewed · Review due · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card for DeepSeek-V3.1, Model ID deepseek.v3.1, raw SHA-256 67c72d5b04fcd77f2acd03d2bb1b7c860432073c5d32e65f8ba852d7ea5863a7, read once on 2026-10-31 under the #837 AWS facts-plus-link basis: "Model lifecycle: Legacy"; "Model EOL date: March 30, 2027"; "Model lifecycle policy: Model lifecycle (For Models Launched Prior to Sept 7 2026)".
  • Amazon Bedrock "Model lifecycle (Legacy)" page (https://docs.aws.amazon.com/bedrock/latest/userguide/model-lifecycle-legacy.html), read once at 2026-10-31T16:36:03Z, raw SHA-256 c33d9d25d7da7f47c098aa864830361de770224f38a790070bada021dac56850: its Legacy table lists DeepSeek-V3.1 (deepseek.v3-v1:0 (bedrock-runtime), deepseek.v3.1 (bedrock-mantle)) with Legacy date September 30, 2026, EOL date March 30, 2027 and public extended access start date "To be announced". The page states: "New customers can't use Legacy models and existing customers may lose access to Legacy models after 15 days of inactivity."
  • That table row's Regions (ap-northeast-1, ap-south-1, ap-southeast-2, ap-southeast-3, eu-north-1, eu-west-2, us-east-2, us-west-2) does not list us-east-1, the only region this offer is published for. The card's own "Model lifecycle: Legacy" field is a statement about the model, so the us-east-1 listing is recorded as deprecated, with this gap noted; whether AWS applies the Legacy state to us-east-1 is not stated.
  • Availability stays available: the us-east-1 price list still lists an on-demand price, and the page says "you can continue to use a Legacy model during this period" (the Legacy period before the EOL date), though not as a new customer. The EOL date is in the future, so no retirement is recorded; reviewAfter is the EOL date, a re-review date, not a retirement.
  • Reviewed by Claude (owner delegation 2026-09-27), #909.

Raw API evidence ↗ · provider API · Claims: bedrock-deepseek-v3-1: pricing · bedrock-deepseek-v3-1: availability · 122 more claims about 61 other catalog entries · Retrieved · Checked · Review due · Confidence: A

  • Raw SHA-256: f1ac6a175fb27b3ad80d2c5d8970be8c2e5f957906349e52d15ec11d43e97896; collector aws-bedrock-price-list-json-1.
  • AWS Price List Bulk API offer file for Amazon Bedrock (us-east-1): offer version 20261111013037, publicationDate 2026-11-11T01:30:37Z.
  • Reviewed on-demand token prices only: each curated model's exact usagetypes in us-east-1, with service_tier "standard" for standard-tier models and with no service_tier attribute for the reviewed untiered models, in US dollars as the offer file states them (hence confidence A). A hosted listing and pricing observation only: no checkpoint, developer, license, context-window, modality or release-date assertions, and the usagetype is not a documented Bedrock model ID.
  • No Anthropic Claude price is published from this offer file: prices are read only from curated usagetypes, and no curated entry is an Anthropic model. (When the collector was reviewed on 2026-10-29, the file listed only legacy Claude models, all without a service tier.)
  • Availability rests on AWS's own offer file listing an on-demand price for the model in us-east-1.

DeepSeek V3.2

Context window: 164K tokens

All rates, conditions and sources

Conditions of every rate below:

  • Listed for model "DeepSeek v3.2" (provider "DeepSeek") in us-east-1 with service_tier "standard", as one USD price dimension from 0 to Inf.
  • AWS lists this price in USD per 1K tokens (unit "1K tokens"); it is multiplied by 1,000 exactly, by decimal arithmetic, to give USD per 1M tokens.
  • Only the standard on-demand tier in us-east-1 is published: flex, priority, batch and global cross-region tiers, other AWS Regions and provisioned throughput are excluded. The offer file lists no standard-tier cache price for this model; missing cache prices are unknown, not zero.

Input: $0.62 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-deepseek.v3.2-mantle-input-tokens-standard.
  • Input tokens.

Output: $1.85 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-deepseek.v3.2-mantle-output-tokens-standard.
  • Output tokens.

Price observed:

API availability: listed as available when checked (not a live service check)

Collection record

Primary source ↗ · provider documentation · Claims: bedrock-deepseek-v3-2: specifications · Retrieved · Checked · Reviewed · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card ("Models at a glance") for DeepSeek V3.2, Model ID deepseek.v3.2 (the offer's price list usagetype stem), raw SHA-256 551dee299adb87077350ac0678a98016700481450025459d75fe015b16cc2f6c, read once on 2026-10-31 under the #837 AWS facts-plus-link basis: "Context window: 164K tokens".
  • Shorthand recorded as stated (#504); not converted to an exact token count. The value is AWS's statement for its Bedrock listing; another seller of the same model may state a different figure.
  • Specifications only: no pricing, availability, capability, lifecycle or release-date assertion.

Raw API evidence ↗ · provider API · Claims: bedrock-deepseek-v3-2: pricing · bedrock-deepseek-v3-2: availability · 122 more claims about 61 other catalog entries · Retrieved · Checked · Review due · Confidence: A

  • Raw SHA-256: f1ac6a175fb27b3ad80d2c5d8970be8c2e5f957906349e52d15ec11d43e97896; collector aws-bedrock-price-list-json-1.
  • AWS Price List Bulk API offer file for Amazon Bedrock (us-east-1): offer version 20261111013037, publicationDate 2026-11-11T01:30:37Z.
  • Reviewed on-demand token prices only: each curated model's exact usagetypes in us-east-1, with service_tier "standard" for standard-tier models and with no service_tier attribute for the reviewed untiered models, in US dollars as the offer file states them (hence confidence A). A hosted listing and pricing observation only: no checkpoint, developer, license, context-window, modality or release-date assertions, and the usagetype is not a documented Bedrock model ID.
  • No Anthropic Claude price is published from this offer file: prices are read only from curated usagetypes, and no curated entry is an Anthropic model. (When the collector was reviewed on 2026-10-29, the file listed only legacy Claude models, all without a service tier.)
  • Availability rests on AWS's own offer file listing an on-demand price for the model in us-east-1.

Devstral 2 123B

Context window: 256K tokens

Deprecated by provider · still listed · price checked Nov 12, 2026 (deprecation notice ↗)

All rates, conditions and sources

Conditions of every rate below:

  • Listed for model "Devstral" (provider "Mistral AI") in us-east-1 with service_tier "standard", as one USD price dimension from 0 to Inf.
  • AWS lists this price in USD per 1K tokens (unit "1K tokens"); it is multiplied by 1,000 exactly, by decimal arithmetic, to give USD per 1M tokens.
  • Only the standard on-demand tier in us-east-1 is published: flex, priority, batch and global cross-region tiers, other AWS Regions and provisioned throughput are excluded. The offer file lists no standard-tier cache price for this model; missing cache prices are unknown, not zero.

Input: $0.40 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-mistral.devstral-2-123b-mantle-input-tokens-standard.
  • Input tokens.

Output: $2.00 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-mistral.devstral-2-123b-mantle-output-tokens-standard.
  • Output tokens.

Price observed:

API availability: listed as available when checked (not a live service check)

Collection record

Primary source ↗ · provider documentation · Claims: bedrock-devstral-2-123b: specifications · Retrieved · Checked · Reviewed · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card ("Models at a glance") for Devstral 2 123B, Model ID mistral.devstral-2-123b (the offer's price list usagetype stem), raw SHA-256 38bf0b6b6c8e88c69cdd25371ac2dff209aeed3a0777b5a8badb080968d1ad70, read once on 2026-10-31 under the #837 AWS facts-plus-link basis: "Context window: 256K tokens".
  • Shorthand recorded as stated (#504); not converted to an exact token count. The value is AWS's statement for its Bedrock listing; another seller of the same model may state a different figure.
  • Specifications only: no pricing, availability, capability, lifecycle or release-date assertion.

Primary source ↗ · provider documentation · Claims: bedrock-devstral-2-123b: availability · Retrieved · Checked · Reviewed · Review due · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card for Devstral 2 123B, Model ID mistral.devstral-2-123b, raw SHA-256 38bf0b6b6c8e88c69cdd25371ac2dff209aeed3a0777b5a8badb080968d1ad70, read once on 2026-10-31 under the #837 AWS facts-plus-link basis: "Model lifecycle: Legacy"; "Model EOL date: March 30, 2027"; "Model lifecycle policy: Model lifecycle (For Models Launched Prior to Sept 7 2026)".
  • Amazon Bedrock "Model lifecycle (Legacy)" page (https://docs.aws.amazon.com/bedrock/latest/userguide/model-lifecycle-legacy.html), read once at 2026-10-31T16:36:03Z, raw SHA-256 c33d9d25d7da7f47c098aa864830361de770224f38a790070bada021dac56850: its Legacy table lists Devstral 2 123B (mistral.devstral-2-123b) with Legacy date September 30, 2026, EOL date March 30, 2027 and public extended access start date "To be announced". The page states: "New customers can't use Legacy models and existing customers may lose access to Legacy models after 15 days of inactivity."
  • That table row's Regions include us-east-1, the region this offer is published for.
  • Availability stays available: the us-east-1 price list still lists an on-demand price, and the page says "you can continue to use a Legacy model during this period" (the Legacy period before the EOL date), though not as a new customer. The EOL date is in the future, so no retirement is recorded; reviewAfter is the EOL date, a re-review date, not a retirement.
  • Reviewed by Claude (owner delegation 2026-09-27), #909.

Raw API evidence ↗ · provider API · Claims: bedrock-devstral-2-123b: pricing · bedrock-devstral-2-123b: availability · 122 more claims about 61 other catalog entries · Retrieved · Checked · Review due · Confidence: A

  • Raw SHA-256: f1ac6a175fb27b3ad80d2c5d8970be8c2e5f957906349e52d15ec11d43e97896; collector aws-bedrock-price-list-json-1.
  • AWS Price List Bulk API offer file for Amazon Bedrock (us-east-1): offer version 20261111013037, publicationDate 2026-11-11T01:30:37Z.
  • Reviewed on-demand token prices only: each curated model's exact usagetypes in us-east-1, with service_tier "standard" for standard-tier models and with no service_tier attribute for the reviewed untiered models, in US dollars as the offer file states them (hence confidence A). A hosted listing and pricing observation only: no checkpoint, developer, license, context-window, modality or release-date assertions, and the usagetype is not a documented Bedrock model ID.
  • No Anthropic Claude price is published from this offer file: prices are read only from curated usagetypes, and no curated entry is an Anthropic model. (When the collector was reviewed on 2026-10-29, the file listed only legacy Claude models, all without a service tier.)
  • Availability rests on AWS's own offer file listing an on-demand price for the model in us-east-1.

Gemma 3 12B

Context window: 128K tokens

Deprecated by provider · still listed · price checked Nov 12, 2026 (deprecation notice ↗)

All rates, conditions and sources

Conditions of every rate below:

  • Listed for model "Gemma 3 12B" (provider "Google") in us-east-1 with service_tier "standard", as one USD price dimension from 0 to Inf.
  • AWS lists this price in USD per 1K tokens (unit "1K tokens"); it is multiplied by 1,000 exactly, by decimal arithmetic, to give USD per 1M tokens.
  • Only the standard on-demand tier in us-east-1 is published: flex, priority, batch and global cross-region tiers, other AWS Regions and provisioned throughput are excluded. The offer file lists no standard-tier cache price for this model; missing cache prices are unknown, not zero.

Input: $0.09 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-google.gemma-3-12b-it-mantle-input-tokens-standard.
  • Input tokens.

Output: $0.29 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-google.gemma-3-12b-it-mantle-output-tokens-standard.
  • Output tokens.

Price observed:

API availability: listed as available when checked (not a live service check)

Collection record

Primary source ↗ · provider documentation · Claims: bedrock-gemma-3-12b-it: specifications · Retrieved · Checked · Reviewed · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card ("Models at a glance") for Gemma 3 12B IT, Model ID google.gemma-3-12b-it (the offer's price list usagetype stem), raw SHA-256 6531c8209b682ab6dc0946cb7a8dd37e0fc81aae9d96a091447bf7c1488c2e94, read once on 2026-10-31 under the #837 AWS facts-plus-link basis: "Context window: 128K tokens".
  • Shorthand recorded as stated (#504); not converted to an exact token count. The value is AWS's statement for its Bedrock listing; another seller of the same model may state a different figure.
  • Specifications only: no pricing, availability, capability, lifecycle or release-date assertion.

Primary source ↗ · provider documentation · Claims: bedrock-gemma-3-12b-it: availability · Retrieved · Checked · Reviewed · Review due · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card for Gemma 3 12B IT, Model ID google.gemma-3-12b-it, raw SHA-256 6531c8209b682ab6dc0946cb7a8dd37e0fc81aae9d96a091447bf7c1488c2e94, read once on 2026-10-31 under the #837 AWS facts-plus-link basis: "Model lifecycle: Legacy"; "Model EOL date: March 30, 2027"; "Model lifecycle policy: Model lifecycle (For Models Launched Prior to Sept 7 2026)".
  • Amazon Bedrock "Model lifecycle (Legacy)" page (https://docs.aws.amazon.com/bedrock/latest/userguide/model-lifecycle-legacy.html), read once at 2026-10-31T16:36:03Z, raw SHA-256 c33d9d25d7da7f47c098aa864830361de770224f38a790070bada021dac56850: its Legacy table lists Gemma 3 12B IT (google.gemma-3-12b-it) with Legacy date September 30, 2026, EOL date March 30, 2027 and public extended access start date "To be announced". The page states: "New customers can't use Legacy models and existing customers may lose access to Legacy models after 15 days of inactivity."
  • That table row's Regions include us-east-1, the region this offer is published for.
  • Availability stays available: the us-east-1 price list still lists an on-demand price, and the page says "you can continue to use a Legacy model during this period" (the Legacy period before the EOL date), though not as a new customer. The EOL date is in the future, so no retirement is recorded; reviewAfter is the EOL date, a re-review date, not a retirement.
  • Reviewed by Claude (owner delegation 2026-09-27), #909.

Raw API evidence ↗ · provider API · Claims: bedrock-gemma-3-12b-it: pricing · bedrock-gemma-3-12b-it: availability · 122 more claims about 61 other catalog entries · Retrieved · Checked · Review due · Confidence: A

  • Raw SHA-256: f1ac6a175fb27b3ad80d2c5d8970be8c2e5f957906349e52d15ec11d43e97896; collector aws-bedrock-price-list-json-1.
  • AWS Price List Bulk API offer file for Amazon Bedrock (us-east-1): offer version 20261111013037, publicationDate 2026-11-11T01:30:37Z.
  • Reviewed on-demand token prices only: each curated model's exact usagetypes in us-east-1, with service_tier "standard" for standard-tier models and with no service_tier attribute for the reviewed untiered models, in US dollars as the offer file states them (hence confidence A). A hosted listing and pricing observation only: no checkpoint, developer, license, context-window, modality or release-date assertions, and the usagetype is not a documented Bedrock model ID.
  • No Anthropic Claude price is published from this offer file: prices are read only from curated usagetypes, and no curated entry is an Anthropic model. (When the collector was reviewed on 2026-10-29, the file listed only legacy Claude models, all without a service tier.)
  • Availability rests on AWS's own offer file listing an on-demand price for the model in us-east-1.

Gemma 3 27B

Context window: 128K tokens

Deprecated by provider · still listed · price checked Nov 12, 2026 (deprecation notice ↗)

All rates, conditions and sources

Conditions of every rate below:

  • Listed for model "Gemma 3 27B" (provider "Google") in us-east-1 with service_tier "standard", as one USD price dimension from 0 to Inf.
  • AWS lists this price in USD per 1K tokens (unit "1K tokens"); it is multiplied by 1,000 exactly, by decimal arithmetic, to give USD per 1M tokens.
  • Only the standard on-demand tier in us-east-1 is published: flex, priority, batch and global cross-region tiers, other AWS Regions and provisioned throughput are excluded. The offer file lists no standard-tier cache price for this model; missing cache prices are unknown, not zero.

Input: $0.23 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-google.gemma-3-27b-it-mantle-input-tokens-standard.
  • Input tokens.

Output: $0.38 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-google.gemma-3-27b-it-mantle-output-tokens-standard.
  • Output tokens.

Price observed:

API availability: listed as available when checked (not a live service check)

Collection record

Primary source ↗ · provider documentation · Claims: bedrock-gemma-3-27b-it: specifications · Retrieved · Checked · Reviewed · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card ("Models at a glance") for Gemma 3 27B PT, Model ID google.gemma-3-27b-it (the offer's price list usagetype stem), raw SHA-256 5a3587cc6dd935200668315ecef8d5f47b9c2bdc6587c1d1351d660cb1521f5b, read once on 2026-10-31 under the #837 AWS facts-plus-link basis: "Context window: 128K tokens". The card is titled "Gemma 3 27B PT", but its Model ID table and code samples name google.gemma-3-27b-it, the price list usagetype stem; the ID is the identity basis.
  • Shorthand recorded as stated (#504); not converted to an exact token count. The value is AWS's statement for its Bedrock listing; another seller of the same model may state a different figure.
  • Specifications only: no pricing, availability, capability, lifecycle or release-date assertion.

Primary source ↗ · provider documentation · Claims: bedrock-gemma-3-27b-it: availability · Retrieved · Checked · Reviewed · Review due · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card for Gemma 3 27B PT, Model ID google.gemma-3-27b-it, raw SHA-256 5a3587cc6dd935200668315ecef8d5f47b9c2bdc6587c1d1351d660cb1521f5b, read once on 2026-10-31 under the #837 AWS facts-plus-link basis: "Model lifecycle: Legacy"; "Model EOL date: March 30, 2027"; "Model lifecycle policy: Model lifecycle (For Models Launched Prior to Sept 7 2026)".
  • Amazon Bedrock "Model lifecycle (Legacy)" page (https://docs.aws.amazon.com/bedrock/latest/userguide/model-lifecycle-legacy.html), read once at 2026-10-31T16:36:03Z, raw SHA-256 c33d9d25d7da7f47c098aa864830361de770224f38a790070bada021dac56850: its Legacy table lists Gemma 3 27B PT (google.gemma-3-27b-it) with Legacy date September 30, 2026, EOL date March 30, 2027 and public extended access start date "To be announced". The page states: "New customers can't use Legacy models and existing customers may lose access to Legacy models after 15 days of inactivity."
  • That table row's Regions include us-east-1, the region this offer is published for.
  • Availability stays available: the us-east-1 price list still lists an on-demand price, and the page says "you can continue to use a Legacy model during this period" (the Legacy period before the EOL date), though not as a new customer. The EOL date is in the future, so no retirement is recorded; reviewAfter is the EOL date, a re-review date, not a retirement.
  • Reviewed by Claude (owner delegation 2026-09-27), #909.

Raw API evidence ↗ · provider API · Claims: bedrock-gemma-3-27b-it: pricing · bedrock-gemma-3-27b-it: availability · 122 more claims about 61 other catalog entries · Retrieved · Checked · Review due · Confidence: A

  • Raw SHA-256: f1ac6a175fb27b3ad80d2c5d8970be8c2e5f957906349e52d15ec11d43e97896; collector aws-bedrock-price-list-json-1.
  • AWS Price List Bulk API offer file for Amazon Bedrock (us-east-1): offer version 20261111013037, publicationDate 2026-11-11T01:30:37Z.
  • Reviewed on-demand token prices only: each curated model's exact usagetypes in us-east-1, with service_tier "standard" for standard-tier models and with no service_tier attribute for the reviewed untiered models, in US dollars as the offer file states them (hence confidence A). A hosted listing and pricing observation only: no checkpoint, developer, license, context-window, modality or release-date assertions, and the usagetype is not a documented Bedrock model ID.
  • No Anthropic Claude price is published from this offer file: prices are read only from curated usagetypes, and no curated entry is an Anthropic model. (When the collector was reviewed on 2026-10-29, the file listed only legacy Claude models, all without a service tier.)
  • Availability rests on AWS's own offer file listing an on-demand price for the model in us-east-1.

Gemma 3 4B

Context window: 128K tokens

Deprecated by provider · still listed · price checked Nov 12, 2026 (deprecation notice ↗)

All rates, conditions and sources

Conditions of every rate below:

  • Listed for model "Gemma 3 4B" (provider "Google") in us-east-1 with service_tier "standard", as one USD price dimension from 0 to Inf.
  • AWS lists this price in USD per 1K tokens (unit "1K tokens"); it is multiplied by 1,000 exactly, by decimal arithmetic, to give USD per 1M tokens.
  • Only the standard on-demand tier in us-east-1 is published: flex, priority, batch and global cross-region tiers, other AWS Regions and provisioned throughput are excluded. The offer file lists no standard-tier cache price for this model; missing cache prices are unknown, not zero.

Input: $0.04 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-google.gemma-3-4b-it-mantle-input-tokens-standard.
  • Input tokens.

Output: $0.08 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-google.gemma-3-4b-it-mantle-output-tokens-standard.
  • Output tokens.

Price observed:

API availability: listed as available when checked (not a live service check)

Collection record

Primary source ↗ · provider documentation · Claims: bedrock-gemma-3-4b-it: specifications · Retrieved · Checked · Reviewed · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card ("Models at a glance") for Gemma 3 4B IT, Model ID google.gemma-3-4b-it (the offer's price list usagetype stem), raw SHA-256 83e032848654959b06a02ab1b5e2524f07c8502a120c127f35d30bf40a109fa5, read once on 2026-10-31 under the #837 AWS facts-plus-link basis: "Context window: 128K tokens".
  • Shorthand recorded as stated (#504); not converted to an exact token count. The value is AWS's statement for its Bedrock listing; another seller of the same model may state a different figure.
  • Specifications only: no pricing, availability, capability, lifecycle or release-date assertion.

Primary source ↗ · provider documentation · Claims: bedrock-gemma-3-4b-it: availability · Retrieved · Checked · Reviewed · Review due · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card for Gemma 3 4B IT, Model ID google.gemma-3-4b-it, raw SHA-256 83e032848654959b06a02ab1b5e2524f07c8502a120c127f35d30bf40a109fa5, read once on 2026-10-31 under the #837 AWS facts-plus-link basis: "Model lifecycle: Legacy"; "Model EOL date: March 30, 2027"; "Model lifecycle policy: Model lifecycle (For Models Launched Prior to Sept 7 2026)".
  • Amazon Bedrock "Model lifecycle (Legacy)" page (https://docs.aws.amazon.com/bedrock/latest/userguide/model-lifecycle-legacy.html), read once at 2026-10-31T16:36:03Z, raw SHA-256 c33d9d25d7da7f47c098aa864830361de770224f38a790070bada021dac56850: its Legacy table lists Gemma 3 4B IT (google.gemma-3-4b-it) with Legacy date September 30, 2026, EOL date March 30, 2027 and public extended access start date "To be announced". The page states: "New customers can't use Legacy models and existing customers may lose access to Legacy models after 15 days of inactivity."
  • That table row's Regions include us-east-1, the region this offer is published for.
  • Availability stays available: the us-east-1 price list still lists an on-demand price, and the page says "you can continue to use a Legacy model during this period" (the Legacy period before the EOL date), though not as a new customer. The EOL date is in the future, so no retirement is recorded; reviewAfter is the EOL date, a re-review date, not a retirement.
  • Reviewed by Claude (owner delegation 2026-09-27), #909.

Raw API evidence ↗ · provider API · Claims: bedrock-gemma-3-4b-it: pricing · bedrock-gemma-3-4b-it: availability · 122 more claims about 61 other catalog entries · Retrieved · Checked · Review due · Confidence: A

  • Raw SHA-256: f1ac6a175fb27b3ad80d2c5d8970be8c2e5f957906349e52d15ec11d43e97896; collector aws-bedrock-price-list-json-1.
  • AWS Price List Bulk API offer file for Amazon Bedrock (us-east-1): offer version 20261111013037, publicationDate 2026-11-11T01:30:37Z.
  • Reviewed on-demand token prices only: each curated model's exact usagetypes in us-east-1, with service_tier "standard" for standard-tier models and with no service_tier attribute for the reviewed untiered models, in US dollars as the offer file states them (hence confidence A). A hosted listing and pricing observation only: no checkpoint, developer, license, context-window, modality or release-date assertions, and the usagetype is not a documented Bedrock model ID.
  • No Anthropic Claude price is published from this offer file: prices are read only from curated usagetypes, and no curated entry is an Anthropic model. (When the collector was reviewed on 2026-10-29, the file listed only legacy Claude models, all without a service tier.)
  • Availability rests on AWS's own offer file listing an on-demand price for the model in us-east-1.

Gemma 4 26B A4B

Context window: 256K tokens

All rates, conditions and sources

Conditions of every rate below:

  • Listed for model "google.gemma-4-26b-a4b" (provider "Google") in us-east-1 with service_tier "standard", as one USD price dimension from 0 to Inf.
  • AWS lists this price in USD per 1K tokens (unit "1K tokens"); it is multiplied by 1,000 exactly, by decimal arithmetic, to give USD per 1M tokens.
  • Only the standard on-demand tier in us-east-1 is published: flex, priority, batch and global cross-region tiers, other AWS Regions and provisioned throughput are excluded. The offer file lists no standard-tier cache price for this model; missing cache prices are unknown, not zero.

Input: $0.13 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-google.gemma-4-26b-a4b-mantle-input-tokens-standard.
  • Input tokens.

Output: $0.40 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-google.gemma-4-26b-a4b-mantle-output-tokens-standard.
  • Output tokens.

Price observed:

API availability: listed as available when checked (not a live service check)

Collection record

Primary source ↗ · provider documentation · Claims: bedrock-gemma-4-26b-a4b: specifications · Retrieved · Checked · Reviewed · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card ("Models at a glance") for Gemma 4 26B-A4B, Model ID google.gemma-4-26b-a4b (the offer's price list usagetype stem), raw SHA-256 46aa00a43a581d4a4c739339a210f7731a85565d8dc4e10b817959a15e2c75c1, read once on 2026-10-31 under the #837 AWS facts-plus-link basis: "Context window: 256K tokens".
  • Shorthand recorded as stated (#504); not converted to an exact token count. The value is AWS's statement for its Bedrock listing; another seller of the same model may state a different figure.
  • Specifications only: no pricing, availability, capability, lifecycle or release-date assertion.

Raw API evidence ↗ · provider API · Claims: bedrock-gemma-4-26b-a4b: pricing · bedrock-gemma-4-26b-a4b: availability · 122 more claims about 61 other catalog entries · Retrieved · Checked · Review due · Confidence: A

  • Raw SHA-256: f1ac6a175fb27b3ad80d2c5d8970be8c2e5f957906349e52d15ec11d43e97896; collector aws-bedrock-price-list-json-1.
  • AWS Price List Bulk API offer file for Amazon Bedrock (us-east-1): offer version 20261111013037, publicationDate 2026-11-11T01:30:37Z.
  • Reviewed on-demand token prices only: each curated model's exact usagetypes in us-east-1, with service_tier "standard" for standard-tier models and with no service_tier attribute for the reviewed untiered models, in US dollars as the offer file states them (hence confidence A). A hosted listing and pricing observation only: no checkpoint, developer, license, context-window, modality or release-date assertions, and the usagetype is not a documented Bedrock model ID.
  • No Anthropic Claude price is published from this offer file: prices are read only from curated usagetypes, and no curated entry is an Anthropic model. (When the collector was reviewed on 2026-10-29, the file listed only legacy Claude models, all without a service tier.)
  • Availability rests on AWS's own offer file listing an on-demand price for the model in us-east-1.

Gemma 4 31B

Context window: 256K tokens

All rates, conditions and sources

Conditions of every rate below:

  • Listed for model "google.gemma-4-31b" (provider "Google") in us-east-1 with service_tier "standard", as one USD price dimension from 0 to Inf.
  • AWS lists this price in USD per 1K tokens (unit "1K tokens"); it is multiplied by 1,000 exactly, by decimal arithmetic, to give USD per 1M tokens.
  • Only the standard on-demand tier in us-east-1 is published: flex, priority, batch and global cross-region tiers, other AWS Regions and provisioned throughput are excluded. The offer file lists no standard-tier cache price for this model; missing cache prices are unknown, not zero.

Input: $0.14 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-google.gemma-4-31b-mantle-input-tokens-standard.
  • Input tokens.

Output: $0.40 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-google.gemma-4-31b-mantle-output-tokens-standard.
  • Output tokens.

Price observed:

API availability: listed as available when checked (not a live service check)

Collection record

Primary source ↗ · provider documentation · Claims: bedrock-gemma-4-31b: specifications · Retrieved · Checked · Reviewed · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card ("Models at a glance") for Gemma 4 31B, Model ID google.gemma-4-31b (the offer's price list usagetype stem), raw SHA-256 1f93b3c0f03d5e48a84246112cc75ad2892a334a0288d0f8470649659973e40e, read once on 2026-10-31 under the #837 AWS facts-plus-link basis: "Context window: 256K tokens".
  • Shorthand recorded as stated (#504); not converted to an exact token count. The value is AWS's statement for its Bedrock listing; another seller of the same model may state a different figure.
  • Specifications only: no pricing, availability, capability, lifecycle or release-date assertion.

Raw API evidence ↗ · provider API · Claims: bedrock-gemma-4-31b: pricing · bedrock-gemma-4-31b: availability · 122 more claims about 61 other catalog entries · Retrieved · Checked · Review due · Confidence: A

  • Raw SHA-256: f1ac6a175fb27b3ad80d2c5d8970be8c2e5f957906349e52d15ec11d43e97896; collector aws-bedrock-price-list-json-1.
  • AWS Price List Bulk API offer file for Amazon Bedrock (us-east-1): offer version 20261111013037, publicationDate 2026-11-11T01:30:37Z.
  • Reviewed on-demand token prices only: each curated model's exact usagetypes in us-east-1, with service_tier "standard" for standard-tier models and with no service_tier attribute for the reviewed untiered models, in US dollars as the offer file states them (hence confidence A). A hosted listing and pricing observation only: no checkpoint, developer, license, context-window, modality or release-date assertions, and the usagetype is not a documented Bedrock model ID.
  • No Anthropic Claude price is published from this offer file: prices are read only from curated usagetypes, and no curated entry is an Anthropic model. (When the collector was reviewed on 2026-10-29, the file listed only legacy Claude models, all without a service tier.)
  • Availability rests on AWS's own offer file listing an on-demand price for the model in us-east-1.

Gemma 4 E2B

Context window: 128K tokens

All rates, conditions and sources

Conditions of every rate below:

  • Listed for model "google.gemma-4-e2b" (provider "Google") in us-east-1 with service_tier "standard", as one USD price dimension from 0 to Inf.
  • AWS lists this price in USD per 1K tokens (unit "1K tokens"); it is multiplied by 1,000 exactly, by decimal arithmetic, to give USD per 1M tokens.
  • Only the standard on-demand tier in us-east-1 is published: flex, priority, batch and global cross-region tiers, other AWS Regions and provisioned throughput are excluded. The offer file lists no standard-tier cache price for this model; missing cache prices are unknown, not zero.

Input: $0.04 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-google.gemma-4-e2b-mantle-input-tokens-standard.
  • Input tokens.

Output: $0.08 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-google.gemma-4-e2b-mantle-output-tokens-standard.
  • Output tokens.

Price observed:

API availability: listed as available when checked (not a live service check)

Collection record

Primary source ↗ · provider documentation · Claims: bedrock-gemma-4-e2b: specifications · Retrieved · Checked · Reviewed · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card ("Models at a glance") for Gemma 4 E2B, Model ID google.gemma-4-e2b (the offer's price list usagetype stem), raw SHA-256 ff5f8e35c71a78d1624eba0f9d75f02ba502e500dfb55228c6659b1c9a9b6926, read once on 2026-10-31 under the #837 AWS facts-plus-link basis: "Context window: 128K tokens".
  • Shorthand recorded as stated (#504); not converted to an exact token count. The value is AWS's statement for its Bedrock listing; another seller of the same model may state a different figure.
  • Specifications only: no pricing, availability, capability, lifecycle or release-date assertion.

Raw API evidence ↗ · provider API · Claims: bedrock-gemma-4-e2b: pricing · bedrock-gemma-4-e2b: availability · 122 more claims about 61 other catalog entries · Retrieved · Checked · Review due · Confidence: A

  • Raw SHA-256: f1ac6a175fb27b3ad80d2c5d8970be8c2e5f957906349e52d15ec11d43e97896; collector aws-bedrock-price-list-json-1.
  • AWS Price List Bulk API offer file for Amazon Bedrock (us-east-1): offer version 20261111013037, publicationDate 2026-11-11T01:30:37Z.
  • Reviewed on-demand token prices only: each curated model's exact usagetypes in us-east-1, with service_tier "standard" for standard-tier models and with no service_tier attribute for the reviewed untiered models, in US dollars as the offer file states them (hence confidence A). A hosted listing and pricing observation only: no checkpoint, developer, license, context-window, modality or release-date assertions, and the usagetype is not a documented Bedrock model ID.
  • No Anthropic Claude price is published from this offer file: prices are read only from curated usagetypes, and no curated entry is an Anthropic model. (When the collector was reviewed on 2026-10-29, the file listed only legacy Claude models, all without a service tier.)
  • Availability rests on AWS's own offer file listing an on-demand price for the model in us-east-1.

GLM 4.7

Context window: 203K tokens

Deprecated by provider · still listed · price checked Nov 12, 2026 (deprecation notice ↗)

All rates, conditions and sources

Conditions of every rate below:

  • Listed for model "GLM 4.7" (provider "Z AI") in us-east-1 with service_tier "standard", as one USD price dimension from 0 to Inf.
  • AWS lists this price in USD per 1K tokens (unit "1K tokens"); it is multiplied by 1,000 exactly, by decimal arithmetic, to give USD per 1M tokens.
  • Only the standard on-demand tier in us-east-1 is published: flex, priority, batch and global cross-region tiers, other AWS Regions and provisioned throughput are excluded. The offer file lists no standard-tier cache price for this model; missing cache prices are unknown, not zero.

Input: $0.60 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-zai.glm-4.7-mantle-input-tokens-standard.
  • Input tokens.

Output: $2.20 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-zai.glm-4.7-mantle-output-tokens-standard.
  • Output tokens.

Price observed:

API availability: listed as available when checked (not a live service check)

Collection record

Primary source ↗ · provider documentation · Claims: bedrock-glm-4-7: specifications · Retrieved · Checked · Reviewed · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card ("Models at a glance") for GLM 4.7, Model ID zai.glm-4.7 (the offer's price list usagetype stem), raw SHA-256 95f57de617f7614e1483412a811e8c22f20f35a7457cb556d9169f3f7ed25e17, read once on 2026-10-31 under the #837 AWS facts-plus-link basis: "Context window: 203K tokens".
  • Shorthand recorded as stated (#504); not converted to an exact token count. The value is AWS's statement for its Bedrock listing; another seller of the same model may state a different figure.
  • Specifications only: no pricing, availability, capability, lifecycle or release-date assertion.

Primary source ↗ · provider documentation · Claims: bedrock-glm-4-7: availability · Retrieved · Checked · Reviewed · Review due · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card for GLM 4.7, Model ID zai.glm-4.7, raw SHA-256 95f57de617f7614e1483412a811e8c22f20f35a7457cb556d9169f3f7ed25e17, read once on 2026-10-31 under the #837 AWS facts-plus-link basis: "Model lifecycle: Legacy"; "Model EOL date: March 30, 2027"; "Model lifecycle policy: Model lifecycle (For Models Launched Prior to Sept 7 2026)".
  • Amazon Bedrock "Model lifecycle (Legacy)" page (https://docs.aws.amazon.com/bedrock/latest/userguide/model-lifecycle-legacy.html), read once at 2026-10-31T16:36:03Z, raw SHA-256 c33d9d25d7da7f47c098aa864830361de770224f38a790070bada021dac56850: its Legacy table lists GLM 4.7 (zai.glm-4.7) with Legacy date September 30, 2026, EOL date March 30, 2027 and public extended access start date "To be announced". The page states: "New customers can't use Legacy models and existing customers may lose access to Legacy models after 15 days of inactivity."
  • That table row's Regions include us-east-1, the region this offer is published for.
  • Availability stays available: the us-east-1 price list still lists an on-demand price, and the page says "you can continue to use a Legacy model during this period" (the Legacy period before the EOL date), though not as a new customer. The EOL date is in the future, so no retirement is recorded; reviewAfter is the EOL date, a re-review date, not a retirement.
  • Reviewed by Claude (owner delegation 2026-09-27), #909.

Raw API evidence ↗ · provider API · Claims: bedrock-glm-4-7: pricing · bedrock-glm-4-7: availability · 122 more claims about 61 other catalog entries · Retrieved · Checked · Review due · Confidence: A

  • Raw SHA-256: f1ac6a175fb27b3ad80d2c5d8970be8c2e5f957906349e52d15ec11d43e97896; collector aws-bedrock-price-list-json-1.
  • AWS Price List Bulk API offer file for Amazon Bedrock (us-east-1): offer version 20261111013037, publicationDate 2026-11-11T01:30:37Z.
  • Reviewed on-demand token prices only: each curated model's exact usagetypes in us-east-1, with service_tier "standard" for standard-tier models and with no service_tier attribute for the reviewed untiered models, in US dollars as the offer file states them (hence confidence A). A hosted listing and pricing observation only: no checkpoint, developer, license, context-window, modality or release-date assertions, and the usagetype is not a documented Bedrock model ID.
  • No Anthropic Claude price is published from this offer file: prices are read only from curated usagetypes, and no curated entry is an Anthropic model. (When the collector was reviewed on 2026-10-29, the file listed only legacy Claude models, all without a service tier.)
  • Availability rests on AWS's own offer file listing an on-demand price for the model in us-east-1.

GLM 4.7 Flash

Context window: 203K tokens

Deprecated by provider · still listed · price checked Nov 12, 2026 (deprecation notice ↗)

All rates, conditions and sources

Conditions of every rate below:

  • Listed for model "GLM 4.7 Flash" (provider "Z AI") in us-east-1 with service_tier "standard", as one USD price dimension from 0 to Inf.
  • AWS lists this price in USD per 1K tokens (unit "1K tokens"); it is multiplied by 1,000 exactly, by decimal arithmetic, to give USD per 1M tokens.
  • Only the standard on-demand tier in us-east-1 is published: flex, priority, batch and global cross-region tiers, other AWS Regions and provisioned throughput are excluded. The offer file lists no standard-tier cache price for this model; missing cache prices are unknown, not zero.

Input: $0.07 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-zai.glm-4.7-flash-mantle-input-tokens-standard.
  • Input tokens.

Output: $0.40 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-zai.glm-4.7-flash-mantle-output-tokens-standard.
  • Output tokens.

Price observed:

API availability: listed as available when checked (not a live service check)

Collection record

Primary source ↗ · provider documentation · Claims: bedrock-glm-4-7-flash: specifications · Retrieved · Checked · Reviewed · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card ("Models at a glance") for GLM 4.7 Flash, Model ID zai.glm-4.7-flash (the offer's price list usagetype stem), raw SHA-256 5dda7a091d9fb63bc78a8b3278ab26d692f61f228045a760d265aac21728501f, read once on 2026-10-31 under the #837 AWS facts-plus-link basis: "Context window: 203K tokens".
  • Shorthand recorded as stated (#504); not converted to an exact token count. The value is AWS's statement for its Bedrock listing; another seller of the same model may state a different figure.
  • Specifications only: no pricing, availability, capability, lifecycle or release-date assertion.

Primary source ↗ · provider documentation · Claims: bedrock-glm-4-7-flash: availability · Retrieved · Checked · Reviewed · Review due · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card for GLM 4.7 Flash, Model ID zai.glm-4.7-flash, raw SHA-256 5dda7a091d9fb63bc78a8b3278ab26d692f61f228045a760d265aac21728501f, read once on 2026-10-31 under the #837 AWS facts-plus-link basis: "Model lifecycle: Legacy"; "Model EOL date: March 30, 2027"; "Model lifecycle policy: Model lifecycle (For Models Launched Prior to Sept 7 2026)".
  • Amazon Bedrock "Model lifecycle (Legacy)" page (https://docs.aws.amazon.com/bedrock/latest/userguide/model-lifecycle-legacy.html), read once at 2026-10-31T16:36:03Z, raw SHA-256 c33d9d25d7da7f47c098aa864830361de770224f38a790070bada021dac56850: its Legacy table lists GLM 4.7 Flash (zai.glm-4.7-flash) with Legacy date September 30, 2026, EOL date March 30, 2027 and public extended access start date "To be announced". The page states: "New customers can't use Legacy models and existing customers may lose access to Legacy models after 15 days of inactivity."
  • That table row's Regions include us-east-1, the region this offer is published for.
  • Availability stays available: the us-east-1 price list still lists an on-demand price, and the page says "you can continue to use a Legacy model during this period" (the Legacy period before the EOL date), though not as a new customer. The EOL date is in the future, so no retirement is recorded; reviewAfter is the EOL date, a re-review date, not a retirement.
  • Reviewed by Claude (owner delegation 2026-09-27), #909.

Raw API evidence ↗ · provider API · Claims: bedrock-glm-4-7-flash: pricing · bedrock-glm-4-7-flash: availability · 122 more claims about 61 other catalog entries · Retrieved · Checked · Review due · Confidence: A

  • Raw SHA-256: f1ac6a175fb27b3ad80d2c5d8970be8c2e5f957906349e52d15ec11d43e97896; collector aws-bedrock-price-list-json-1.
  • AWS Price List Bulk API offer file for Amazon Bedrock (us-east-1): offer version 20261111013037, publicationDate 2026-11-11T01:30:37Z.
  • Reviewed on-demand token prices only: each curated model's exact usagetypes in us-east-1, with service_tier "standard" for standard-tier models and with no service_tier attribute for the reviewed untiered models, in US dollars as the offer file states them (hence confidence A). A hosted listing and pricing observation only: no checkpoint, developer, license, context-window, modality or release-date assertions, and the usagetype is not a documented Bedrock model ID.
  • No Anthropic Claude price is published from this offer file: prices are read only from curated usagetypes, and no curated entry is an Anthropic model. (When the collector was reviewed on 2026-10-29, the file listed only legacy Claude models, all without a service tier.)
  • Availability rests on AWS's own offer file listing an on-demand price for the model in us-east-1.

GLM 5

Context window: 200K tokens

All rates, conditions and sources

Conditions of every rate below:

  • Listed for model "GLM 5" (provider "Z AI") in us-east-1 with service_tier "standard", as one USD price dimension from 0 to Inf.
  • AWS lists this price in USD per 1K tokens (unit "1K tokens"); it is multiplied by 1,000 exactly, by decimal arithmetic, to give USD per 1M tokens.
  • Only the standard on-demand tier in us-east-1 is published: flex, priority, batch and global cross-region tiers, other AWS Regions and provisioned throughput are excluded. The offer file lists no standard-tier cache price for this model; missing cache prices are unknown, not zero.

Input: $1.00 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-zai.glm5-mantle-input-tokens-standard.
  • Input tokens.

Output: $3.20 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-zai.glm5-mantle-output-tokens-standard.
  • Output tokens.

Capability (Epoch AI): ECI 146.5 (90% range 145.0–148.1) · Epoch AI score for GLM-5, model dated Feb 11, 2026

Price observed:

API availability: listed as available when checked (not a live service check)

Collection record

Primary source ↗ · provider documentation · Claims: bedrock-glm-5: specifications · Retrieved · Checked · Reviewed · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card ("Models at a glance") for GLM 5, Model ID zai.glm-5; the price list usagetype stem is zai.glm5, which the card does not state, so identity rests on the card title "GLM 5" and the price list's model "GLM 5" (provider "Z AI"), the only GLM 5 card in the index (GLM 5.3 has its own), raw SHA-256 1fe53f68bbe1073236f9305dca6c9c3d64bcfe8868fad0426fa7f0f8d70e662b, read once on 2026-10-31 under the #837 AWS facts-plus-link basis: "Context window: 200K tokens".
  • Shorthand recorded as stated (#504); not converted to an exact token count. The value is AWS's statement for its Bedrock listing; another seller of the same model may state a different figure.
  • Specifications only: no pricing, availability, capability, lifecycle or release-date assertion.

Raw API evidence ↗ · provider API · Claims: bedrock-glm-5: pricing · bedrock-glm-5: availability · 122 more claims about 61 other catalog entries · Retrieved · Checked · Review due · Confidence: A

  • Raw SHA-256: f1ac6a175fb27b3ad80d2c5d8970be8c2e5f957906349e52d15ec11d43e97896; collector aws-bedrock-price-list-json-1.
  • AWS Price List Bulk API offer file for Amazon Bedrock (us-east-1): offer version 20261111013037, publicationDate 2026-11-11T01:30:37Z.
  • Reviewed on-demand token prices only: each curated model's exact usagetypes in us-east-1, with service_tier "standard" for standard-tier models and with no service_tier attribute for the reviewed untiered models, in US dollars as the offer file states them (hence confidence A). A hosted listing and pricing observation only: no checkpoint, developer, license, context-window, modality or release-date assertions, and the usagetype is not a documented Bedrock model ID.
  • No Anthropic Claude price is published from this offer file: prices are read only from curated usagetypes, and no curated entry is an Anthropic model. (When the collector was reviewed on 2026-10-29, the file listed only legacy Claude models, all without a service tier.)
  • Availability rests on AWS's own offer file listing an on-demand price for the model in us-east-1.

GPT OSS Safeguard 120B

Context window: 128K tokens

All rates, conditions and sources

Conditions of every rate below:

  • Listed for model "GPT OSS Safeguard 120B" (provider "OpenAI") in us-east-1 with service_tier "standard", as one USD price dimension from 0 to Inf.
  • AWS lists this price in USD per 1K tokens (unit "1K tokens"); it is multiplied by 1,000 exactly, by decimal arithmetic, to give USD per 1M tokens.
  • Only the standard on-demand tier in us-east-1 is published: flex, priority, batch and global cross-region tiers, other AWS Regions and provisioned throughput are excluded. The offer file lists no standard-tier cache price for this model; missing cache prices are unknown, not zero.

Input: $0.15 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-openai.gpt-oss-safeguard-120b-mantle-input-tokens-standard.
  • Input tokens.

Output: $0.60 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-openai.gpt-oss-safeguard-120b-mantle-output-tokens-standard.
  • Output tokens.

Price observed:

API availability: listed as available when checked (not a live service check)

Collection record

Primary source ↗ · provider documentation · Claims: bedrock-gpt-oss-safeguard-120b: specifications · Retrieved · Checked · Reviewed · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card ("Models at a glance") for GPT OSS Safeguard 120B, Model ID openai.gpt-oss-safeguard-120b (the offer's price list usagetype stem), raw SHA-256 fd3929def03bf9230b89e98c1be144525044c9fbbead5841b2a93d0d8a002706, read once on 2026-10-31 under the #837 AWS facts-plus-link basis: "Context window: 128K tokens".
  • Shorthand recorded as stated (#504); not converted to an exact token count. The value is AWS's statement for its Bedrock listing; another seller of the same model may state a different figure.
  • Specifications only: no pricing, availability, capability, lifecycle or release-date assertion.

Raw API evidence ↗ · provider API · Claims: bedrock-gpt-oss-safeguard-120b: pricing · bedrock-gpt-oss-safeguard-120b: availability · 122 more claims about 61 other catalog entries · Retrieved · Checked · Review due · Confidence: A

  • Raw SHA-256: f1ac6a175fb27b3ad80d2c5d8970be8c2e5f957906349e52d15ec11d43e97896; collector aws-bedrock-price-list-json-1.
  • AWS Price List Bulk API offer file for Amazon Bedrock (us-east-1): offer version 20261111013037, publicationDate 2026-11-11T01:30:37Z.
  • Reviewed on-demand token prices only: each curated model's exact usagetypes in us-east-1, with service_tier "standard" for standard-tier models and with no service_tier attribute for the reviewed untiered models, in US dollars as the offer file states them (hence confidence A). A hosted listing and pricing observation only: no checkpoint, developer, license, context-window, modality or release-date assertions, and the usagetype is not a documented Bedrock model ID.
  • No Anthropic Claude price is published from this offer file: prices are read only from curated usagetypes, and no curated entry is an Anthropic model. (When the collector was reviewed on 2026-10-29, the file listed only legacy Claude models, all without a service tier.)
  • Availability rests on AWS's own offer file listing an on-demand price for the model in us-east-1.

GPT OSS Safeguard 20B

Context window: 128K tokens

All rates, conditions and sources

Conditions of every rate below:

  • Listed for model "GPT OSS Safeguard 20B" (provider "OpenAI") in us-east-1 with service_tier "standard", as one USD price dimension from 0 to Inf.
  • AWS lists this price in USD per 1K tokens (unit "1K tokens"); it is multiplied by 1,000 exactly, by decimal arithmetic, to give USD per 1M tokens.
  • Only the standard on-demand tier in us-east-1 is published: flex, priority, batch and global cross-region tiers, other AWS Regions and provisioned throughput are excluded. The offer file lists no standard-tier cache price for this model; missing cache prices are unknown, not zero.

Input: $0.07 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-openai.gpt-oss-safeguard-20b-mantle-input-tokens-standard.
  • Input tokens.

Output: $0.20 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-openai.gpt-oss-safeguard-20b-mantle-output-tokens-standard.
  • Output tokens.

Price observed:

API availability: listed as available when checked (not a live service check)

Collection record

Primary source ↗ · provider documentation · Claims: bedrock-gpt-oss-safeguard-20b: specifications · Retrieved · Checked · Reviewed · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card ("Models at a glance") for GPT OSS Safeguard 20B, Model ID openai.gpt-oss-safeguard-20b (the offer's price list usagetype stem), raw SHA-256 dbd336ce0d72757d99af94d2af6da0411408296057e3b6a7dcae7be7d55719a8, read once on 2026-10-31 under the #837 AWS facts-plus-link basis: "Context window: 128K tokens".
  • Shorthand recorded as stated (#504); not converted to an exact token count. The value is AWS's statement for its Bedrock listing; another seller of the same model may state a different figure.
  • Specifications only: no pricing, availability, capability, lifecycle or release-date assertion.

Raw API evidence ↗ · provider API · Claims: bedrock-gpt-oss-safeguard-20b: pricing · bedrock-gpt-oss-safeguard-20b: availability · 122 more claims about 61 other catalog entries · Retrieved · Checked · Review due · Confidence: A

  • Raw SHA-256: f1ac6a175fb27b3ad80d2c5d8970be8c2e5f957906349e52d15ec11d43e97896; collector aws-bedrock-price-list-json-1.
  • AWS Price List Bulk API offer file for Amazon Bedrock (us-east-1): offer version 20261111013037, publicationDate 2026-11-11T01:30:37Z.
  • Reviewed on-demand token prices only: each curated model's exact usagetypes in us-east-1, with service_tier "standard" for standard-tier models and with no service_tier attribute for the reviewed untiered models, in US dollars as the offer file states them (hence confidence A). A hosted listing and pricing observation only: no checkpoint, developer, license, context-window, modality or release-date assertions, and the usagetype is not a documented Bedrock model ID.
  • No Anthropic Claude price is published from this offer file: prices are read only from curated usagetypes, and no curated entry is an Anthropic model. (When the collector was reviewed on 2026-10-29, the file listed only legacy Claude models, all without a service tier.)
  • Availability rests on AWS's own offer file listing an on-demand price for the model in us-east-1.

gpt-oss-120b

Context window: 128K tokens

All rates, conditions and sources

Conditions of every rate below:

  • Listed for model "gpt-oss-120b" (provider "OpenAI") in us-east-1 with service_tier "standard", as one USD price dimension from 0 to Inf.
  • AWS lists this price in USD per 1K tokens (unit "1K tokens"); it is multiplied by 1,000 exactly, by decimal arithmetic, to give USD per 1M tokens.
  • Only the standard on-demand tier in us-east-1 is published: flex, priority, batch and global cross-region tiers, other AWS Regions and provisioned throughput are excluded. The offer file lists no standard-tier cache price for this model; missing cache prices are unknown, not zero.

Input: $0.15 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-openai.gpt-oss-120b-mantle-input-tokens-standard.
  • Input tokens.

Output: $0.60 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-openai.gpt-oss-120b-mantle-output-tokens-standard.
  • Output tokens.

Capability (Epoch AI): ECI 140.2 (90% range 138.3–142.1) · range spans bands · Epoch AI score for gpt-oss-120b, model dated Aug 5, 2025

Price observed:

API availability: listed as available when checked (not a live service check)

Collection record

Primary source ↗ · provider documentation · Claims: bedrock-gpt-oss-120b: specifications · Retrieved · Checked · Reviewed · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card ("Models at a glance") for gpt-oss-120b, Model ID openai.gpt-oss-120b (the offer's price list usagetype stem), raw SHA-256 e8362d2487b7637bdd398b1115a3dcb257cb835c72934f0b981df6112b6e8316, read once on 2026-10-31 under the #837 AWS facts-plus-link basis: "Context window: 128K tokens".
  • Shorthand recorded as stated (#504); not converted to an exact token count. The value is AWS's statement for its Bedrock listing; another seller of the same model may state a different figure.
  • Specifications only: no pricing, availability, capability, lifecycle or release-date assertion.

Raw API evidence ↗ · provider API · Claims: bedrock-gpt-oss-120b: pricing · bedrock-gpt-oss-120b: availability · 122 more claims about 61 other catalog entries · Retrieved · Checked · Review due · Confidence: A

  • Raw SHA-256: f1ac6a175fb27b3ad80d2c5d8970be8c2e5f957906349e52d15ec11d43e97896; collector aws-bedrock-price-list-json-1.
  • AWS Price List Bulk API offer file for Amazon Bedrock (us-east-1): offer version 20261111013037, publicationDate 2026-11-11T01:30:37Z.
  • Reviewed on-demand token prices only: each curated model's exact usagetypes in us-east-1, with service_tier "standard" for standard-tier models and with no service_tier attribute for the reviewed untiered models, in US dollars as the offer file states them (hence confidence A). A hosted listing and pricing observation only: no checkpoint, developer, license, context-window, modality or release-date assertions, and the usagetype is not a documented Bedrock model ID.
  • No Anthropic Claude price is published from this offer file: prices are read only from curated usagetypes, and no curated entry is an Anthropic model. (When the collector was reviewed on 2026-10-29, the file listed only legacy Claude models, all without a service tier.)
  • Availability rests on AWS's own offer file listing an on-demand price for the model in us-east-1.

gpt-oss-20b

Context window: 128K tokens

All rates, conditions and sources

Conditions of every rate below:

  • Listed for model "gpt-oss-20b" (provider "OpenAI") in us-east-1 with service_tier "standard", as one USD price dimension from 0 to Inf.
  • AWS lists this price in USD per 1K tokens (unit "1K tokens"); it is multiplied by 1,000 exactly, by decimal arithmetic, to give USD per 1M tokens.
  • Only the standard on-demand tier in us-east-1 is published: flex, priority, batch and global cross-region tiers, other AWS Regions and provisioned throughput are excluded. The offer file lists no standard-tier cache price for this model; missing cache prices are unknown, not zero.

Input: $0.07 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-openai.gpt-oss-20b-mantle-input-tokens-standard.
  • Input tokens.

Output: $0.30 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-openai.gpt-oss-20b-mantle-output-tokens-standard.
  • Output tokens.

Capability (Epoch AI): ECI 134.9 (90% range 132.3–137.3) · Epoch AI score for gpt-oss-20b, model dated Aug 5, 2025

Price observed:

API availability: listed as available when checked (not a live service check)

Collection record

Primary source ↗ · provider documentation · Claims: bedrock-gpt-oss-20b: specifications · Retrieved · Checked · Reviewed · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card ("Models at a glance") for gpt-oss-20b, Model ID openai.gpt-oss-20b (the offer's price list usagetype stem), raw SHA-256 a7a220b6600f8b9d7d9bc9898db5f793bacbe93f59dbaf3279665a074469b360, read once on 2026-10-31 under the #837 AWS facts-plus-link basis: "Context window: 128K tokens".
  • Shorthand recorded as stated (#504); not converted to an exact token count. The value is AWS's statement for its Bedrock listing; another seller of the same model may state a different figure.
  • Specifications only: no pricing, availability, capability, lifecycle or release-date assertion.

Raw API evidence ↗ · provider API · Claims: bedrock-gpt-oss-20b: pricing · bedrock-gpt-oss-20b: availability · 122 more claims about 61 other catalog entries · Retrieved · Checked · Review due · Confidence: A

  • Raw SHA-256: f1ac6a175fb27b3ad80d2c5d8970be8c2e5f957906349e52d15ec11d43e97896; collector aws-bedrock-price-list-json-1.
  • AWS Price List Bulk API offer file for Amazon Bedrock (us-east-1): offer version 20261111013037, publicationDate 2026-11-11T01:30:37Z.
  • Reviewed on-demand token prices only: each curated model's exact usagetypes in us-east-1, with service_tier "standard" for standard-tier models and with no service_tier attribute for the reviewed untiered models, in US dollars as the offer file states them (hence confidence A). A hosted listing and pricing observation only: no checkpoint, developer, license, context-window, modality or release-date assertions, and the usagetype is not a documented Bedrock model ID.
  • No Anthropic Claude price is published from this offer file: prices are read only from curated usagetypes, and no curated entry is an Anthropic model. (When the collector was reviewed on 2026-10-29, the file listed only legacy Claude models, all without a service tier.)
  • Availability rests on AWS's own offer file listing an on-demand price for the model in us-east-1.

Grok 4.3

Context window: 1M tokens

All rates, conditions and sources

Conditions of every rate below:

  • Listed for model "xai.grok-4.3" in us-east-1 with service_tier "standard", as one USD price dimension from 0 to Inf.
  • AWS lists this price in USD per 1M tokens (unit "1M tokens"); it is published as listed.
  • Only the standard on-demand tier in us-east-1 is published: flex, priority, batch and global cross-region tiers, other AWS Regions and provisioned throughput are excluded. The offer file lists cache prices for this model as separate usagetypes.

Input: $1.25 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-xai.grok-4.3-mantle-input-tokens-standard.
  • Input tokens.

Output: $2.50 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-xai.grok-4.3-mantle-output-tokens-standard.
  • Output tokens.

Cache read: $0.20 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-xai.grok-4.3-mantle-cache-read-tokens-standard.
  • Cache read tokens (a "cache-read-tokens" usagetype). The offer file states no cache eligibility or cache lifetime for reads.

Price observed:

API availability: listed as available when checked (not a live service check)

Collection record

Primary source ↗ · provider documentation · Claims: bedrock-grok-4-3: specifications · Retrieved · Checked · Reviewed · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card ("Models at a glance") for Grok 4.3, Model ID xai.grok-4.3 (the offer's price list usagetype stem), raw SHA-256 bffc85a1f39e987fc5e841751f2984f8ff2838607d5c2a59cd5908dfc401556c, read once on 2026-10-31 under the #837 AWS facts-plus-link basis: "Context window: 1M tokens".
  • Shorthand recorded as stated (#504); not converted to an exact token count. The value is AWS's statement for its Bedrock listing; another seller of the same model may state a different figure.
  • Specifications only: no pricing, availability, capability, lifecycle or release-date assertion.

Raw API evidence ↗ · provider API · Claims: bedrock-grok-4-3: pricing · bedrock-grok-4-3: availability · 122 more claims about 61 other catalog entries · Retrieved · Checked · Review due · Confidence: A

  • Raw SHA-256: f1ac6a175fb27b3ad80d2c5d8970be8c2e5f957906349e52d15ec11d43e97896; collector aws-bedrock-price-list-json-1.
  • AWS Price List Bulk API offer file for Amazon Bedrock (us-east-1): offer version 20261111013037, publicationDate 2026-11-11T01:30:37Z.
  • Reviewed on-demand token prices only: each curated model's exact usagetypes in us-east-1, with service_tier "standard" for standard-tier models and with no service_tier attribute for the reviewed untiered models, in US dollars as the offer file states them (hence confidence A). A hosted listing and pricing observation only: no checkpoint, developer, license, context-window, modality or release-date assertions, and the usagetype is not a documented Bedrock model ID.
  • No Anthropic Claude price is published from this offer file: prices are read only from curated usagetypes, and no curated entry is an Anthropic model. (When the collector was reviewed on 2026-10-29, the file listed only legacy Claude models, all without a service tier.)
  • Availability rests on AWS's own offer file listing an on-demand price for the model in us-east-1.

Grok 4.6

Context window: 500K tokens

All rates, conditions and sources

Conditions of every rate below:

  • Listed for model "xai.grok-4.6" in us-east-1 with service_tier "standard", as one USD price dimension from 0 to Inf.
  • AWS lists this price in USD per 1K tokens (unit "1K tokens"); it is multiplied by 1,000 exactly, by decimal arithmetic, to give USD per 1M tokens.
  • Only the standard on-demand tier in us-east-1 is published: flex, priority, batch and global cross-region tiers, other AWS Regions and provisioned throughput are excluded. The offer file lists cache prices for this model as separate usagetypes.

Input: $2.20 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-xai.grok-4.6-mantle-input-tokens-standard.
  • Input tokens.

Output: $6.60 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-xai.grok-4.6-mantle-output-tokens-standard.
  • Output tokens.

Cache read: $0.55 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-xai.grok-4.6-mantle-cache-read-tokens-standard.
  • Cache read tokens (a "cache-read-tokens" usagetype). The offer file states no cache eligibility or cache lifetime for reads.

Price observed:

API availability: listed as available when checked (not a live service check)

Collection record

Primary source ↗ · provider documentation · Claims: bedrock-grok-4-6: specifications · Retrieved · Checked · Reviewed · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card ("Models at a glance") for Grok 4.6, Model ID xai.grok-4.6 (the offer's price list usagetype stem), raw SHA-256 9462afa12617eaece25ef815d9f20687ed0be16e8a3e057bed836420f2e80a3f, read once on 2026-10-31 under the #837 AWS facts-plus-link basis: "Context window: 500K tokens".
  • Shorthand recorded as stated (#504); not converted to an exact token count. The value is AWS's statement for its Bedrock listing; another seller of the same model may state a different figure.
  • Specifications only: no pricing, availability, capability, lifecycle or release-date assertion.

Raw API evidence ↗ · provider API · Claims: bedrock-grok-4-6: pricing · bedrock-grok-4-6: availability · 122 more claims about 61 other catalog entries · Retrieved · Checked · Review due · Confidence: A

  • Raw SHA-256: f1ac6a175fb27b3ad80d2c5d8970be8c2e5f957906349e52d15ec11d43e97896; collector aws-bedrock-price-list-json-1.
  • AWS Price List Bulk API offer file for Amazon Bedrock (us-east-1): offer version 20261111013037, publicationDate 2026-11-11T01:30:37Z.
  • Reviewed on-demand token prices only: each curated model's exact usagetypes in us-east-1, with service_tier "standard" for standard-tier models and with no service_tier attribute for the reviewed untiered models, in US dollars as the offer file states them (hence confidence A). A hosted listing and pricing observation only: no checkpoint, developer, license, context-window, modality or release-date assertions, and the usagetype is not a documented Bedrock model ID.
  • No Anthropic Claude price is published from this offer file: prices are read only from curated usagetypes, and no curated entry is an Anthropic model. (When the collector was reviewed on 2026-10-29, the file listed only legacy Claude models, all without a service tier.)
  • Availability rests on AWS's own offer file listing an on-demand price for the model in us-east-1.

Grok 4.7

Context window: 524K tokens

All rates, conditions and sources

Conditions of every rate below:

  • Listed for model "Grok 4.7" (provider "xAI") in us-east-1 with service_tier "standard", as one USD price dimension from 0 to Inf.
  • AWS lists this price in USD per 1K tokens (unit "1K tokens"); it is multiplied by 1,000 exactly, by decimal arithmetic, to give USD per 1M tokens.
  • Only the standard on-demand tier in us-east-1 is published: flex, priority, batch and global cross-region tiers, other AWS Regions and provisioned throughput are excluded. The offer file lists cache prices for this model as separate usagetypes.

Input: $2.20 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-xai.grok-4.7-mantle-input-tokens-standard.
  • Input tokens.

Output: $6.60 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-xai.grok-4.7-mantle-output-tokens-standard.
  • Output tokens.

Cache read: $0.55 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-xai.grok-4.7-mantle-cache-read-tokens-standard.
  • Cache read tokens (a "cache-read-tokens" usagetype). The offer file states no cache eligibility or cache lifetime for reads.

Capability (Epoch AI): ECI 156.4 (90% range 154.5–158.5) · Epoch AI score for Grok 4.7, model dated Sep 21, 2026

Context window: 524,288 tokens

Price observed:

API availability: listed as available when checked (not a live service check)

Collection record

Primary source ↗ · provider documentation · Claims: bedrock-grok-4-7: specifications · Retrieved · Checked · Reviewed · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card ("Models at a glance") for Grok 4.7, Model ID xai.grok-4.7 (the offer's price list usagetype stem), raw SHA-256 e71fc77d877a3221d98ffd410048494af0b1a8f8eb2611c1a23dd228a704a69c, read once on 2026-10-31 under the #837 AWS facts-plus-link basis: "Context window: 524,288 tokens, shared between input and output".
  • Exact figure recorded as stated (524,288 tokens, input and output combined). The value is AWS's statement for its Bedrock listing; another seller of the same model may state a different figure.
  • Specifications only: no pricing, availability, capability, lifecycle or release-date assertion.

Raw API evidence ↗ · provider API · Claims: bedrock-grok-4-7: pricing · bedrock-grok-4-7: availability · 122 more claims about 61 other catalog entries · Retrieved · Checked · Review due · Confidence: A

  • Raw SHA-256: f1ac6a175fb27b3ad80d2c5d8970be8c2e5f957906349e52d15ec11d43e97896; collector aws-bedrock-price-list-json-1.
  • AWS Price List Bulk API offer file for Amazon Bedrock (us-east-1): offer version 20261111013037, publicationDate 2026-11-11T01:30:37Z.
  • Reviewed on-demand token prices only: each curated model's exact usagetypes in us-east-1, with service_tier "standard" for standard-tier models and with no service_tier attribute for the reviewed untiered models, in US dollars as the offer file states them (hence confidence A). A hosted listing and pricing observation only: no checkpoint, developer, license, context-window, modality or release-date assertions, and the usagetype is not a documented Bedrock model ID.
  • No Anthropic Claude price is published from this offer file: prices are read only from curated usagetypes, and no curated entry is an Anthropic model. (When the collector was reviewed on 2026-10-29, the file listed only legacy Claude models, all without a service tier.)
  • Availability rests on AWS's own offer file listing an on-demand price for the model in us-east-1.

Kimi K2 Thinking

Context window: 256K tokens

Deprecated by provider · still listed · price checked Nov 12, 2026 (deprecation notice ↗)

All rates, conditions and sources

Conditions of every rate below:

  • Listed for model "Kimi K2 Thinking" (provider "Moonshot AI") in us-east-1 with service_tier "standard", as one USD price dimension from 0 to Inf.
  • AWS lists this price in USD per 1K tokens (unit "1K tokens"); it is multiplied by 1,000 exactly, by decimal arithmetic, to give USD per 1M tokens.
  • Only the standard on-demand tier in us-east-1 is published: flex, priority, batch and global cross-region tiers, other AWS Regions and provisioned throughput are excluded. The offer file lists no standard-tier cache price for this model; missing cache prices are unknown, not zero.

Input: $0.60 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-moonshotai.kimi-k2-thinking-mantle-input-tokens-standard.
  • Input tokens.

Output: $2.50 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-moonshotai.kimi-k2-thinking-mantle-output-tokens-standard.
  • Output tokens.

Price observed:

API availability: listed as available when checked (not a live service check)

Collection record

Primary source ↗ · provider documentation · Claims: bedrock-kimi-k2-thinking: specifications · Retrieved · Checked · Reviewed · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card ("Models at a glance") for Kimi K2 Thinking, Model ID moonshotai.kimi-k2-thinking (the offer's price list usagetype stem), raw SHA-256 38d5c7509183b44b1317fc99f3dd427a48c4570c525b8bd5664b77b32b55db6b, read once on 2026-10-31 under the #837 AWS facts-plus-link basis: "Context window: 256K tokens".
  • Shorthand recorded as stated (#504); not converted to an exact token count. The value is AWS's statement for its Bedrock listing; another seller of the same model may state a different figure.
  • Specifications only: no pricing, availability, capability, lifecycle or release-date assertion.

Primary source ↗ · provider documentation · Claims: bedrock-kimi-k2-thinking: availability · Retrieved · Checked · Reviewed · Review due · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card for Kimi K2 Thinking, Model ID moonshotai.kimi-k2-thinking, raw SHA-256 38d5c7509183b44b1317fc99f3dd427a48c4570c525b8bd5664b77b32b55db6b, read once on 2026-10-31 under the #837 AWS facts-plus-link basis: "Model lifecycle: Legacy"; "Model EOL date: March 30, 2027"; "Model lifecycle policy: Model lifecycle (For Models Launched Prior to Sept 7 2026)".
  • Amazon Bedrock "Model lifecycle (Legacy)" page (https://docs.aws.amazon.com/bedrock/latest/userguide/model-lifecycle-legacy.html), read once at 2026-10-31T16:36:03Z, raw SHA-256 c33d9d25d7da7f47c098aa864830361de770224f38a790070bada021dac56850: its Legacy table lists Kimi K2 Thinking (moonshot.kimi-k2-thinking (bedrock-runtime), moonshotai.kimi-k2-thinking (bedrock-mantle)) with Legacy date September 30, 2026, EOL date March 30, 2027 and public extended access start date "To be announced". The page states: "New customers can't use Legacy models and existing customers may lose access to Legacy models after 15 days of inactivity."
  • That table row's Regions include us-east-1, the region this offer is published for.
  • Availability stays available: the us-east-1 price list still lists an on-demand price, and the page says "you can continue to use a Legacy model during this period" (the Legacy period before the EOL date), though not as a new customer. The EOL date is in the future, so no retirement is recorded; reviewAfter is the EOL date, a re-review date, not a retirement.
  • Reviewed by Claude (owner delegation 2026-09-27), #909.

Raw API evidence ↗ · provider API · Claims: bedrock-kimi-k2-thinking: pricing · bedrock-kimi-k2-thinking: availability · 122 more claims about 61 other catalog entries · Retrieved · Checked · Review due · Confidence: A

  • Raw SHA-256: f1ac6a175fb27b3ad80d2c5d8970be8c2e5f957906349e52d15ec11d43e97896; collector aws-bedrock-price-list-json-1.
  • AWS Price List Bulk API offer file for Amazon Bedrock (us-east-1): offer version 20261111013037, publicationDate 2026-11-11T01:30:37Z.
  • Reviewed on-demand token prices only: each curated model's exact usagetypes in us-east-1, with service_tier "standard" for standard-tier models and with no service_tier attribute for the reviewed untiered models, in US dollars as the offer file states them (hence confidence A). A hosted listing and pricing observation only: no checkpoint, developer, license, context-window, modality or release-date assertions, and the usagetype is not a documented Bedrock model ID.
  • No Anthropic Claude price is published from this offer file: prices are read only from curated usagetypes, and no curated entry is an Anthropic model. (When the collector was reviewed on 2026-10-29, the file listed only legacy Claude models, all without a service tier.)
  • Availability rests on AWS's own offer file listing an on-demand price for the model in us-east-1.

Kimi K2.5

Context window: 256K tokens

All rates, conditions and sources

Conditions of every rate below:

  • Listed for model "Kimi K2.5" (provider "Moonshot AI") in us-east-1 with service_tier "standard", as one USD price dimension from 0 to Inf.
  • AWS lists this price in USD per 1K tokens (unit "1K tokens"); it is multiplied by 1,000 exactly, by decimal arithmetic, to give USD per 1M tokens.
  • Only the standard on-demand tier in us-east-1 is published: flex, priority, batch and global cross-region tiers, other AWS Regions and provisioned throughput are excluded. The offer file lists no standard-tier cache price for this model; missing cache prices are unknown, not zero.

Input: $0.60 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-moonshotai.kimi-k2.5-mantle-input-tokens-standard.
  • Input tokens.

Output: $3.00 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-moonshotai.kimi-k2.5-mantle-output-tokens-standard.
  • Output tokens.

Capability (Epoch AI): ECI 148.8 (90% range 147.3–150.4) · range spans bands · Epoch AI score for Kimi K2.5, model dated Jan 27, 2026

Price observed:

API availability: listed as available when checked (not a live service check)

Collection record

Primary source ↗ · provider documentation · Claims: bedrock-kimi-k2-5: specifications · Retrieved · Checked · Reviewed · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card ("Models at a glance") for Kimi K2.5, Model ID moonshotai.kimi-k2.5 (the offer's price list usagetype stem), raw SHA-256 d70a1cfaff30cf7cc14e428f3b93b087acf41a8c0827490a594250a67da8fcfc, read once on 2026-10-31 under the #837 AWS facts-plus-link basis: "Context window: 256K tokens".
  • Shorthand recorded as stated (#504); not converted to an exact token count. The value is AWS's statement for its Bedrock listing; another seller of the same model may state a different figure.
  • Specifications only: no pricing, availability, capability, lifecycle or release-date assertion.

Raw API evidence ↗ · provider API · Claims: bedrock-kimi-k2-5: pricing · bedrock-kimi-k2-5: availability · 122 more claims about 61 other catalog entries · Retrieved · Checked · Review due · Confidence: A

  • Raw SHA-256: f1ac6a175fb27b3ad80d2c5d8970be8c2e5f957906349e52d15ec11d43e97896; collector aws-bedrock-price-list-json-1.
  • AWS Price List Bulk API offer file for Amazon Bedrock (us-east-1): offer version 20261111013037, publicationDate 2026-11-11T01:30:37Z.
  • Reviewed on-demand token prices only: each curated model's exact usagetypes in us-east-1, with service_tier "standard" for standard-tier models and with no service_tier attribute for the reviewed untiered models, in US dollars as the offer file states them (hence confidence A). A hosted listing and pricing observation only: no checkpoint, developer, license, context-window, modality or release-date assertions, and the usagetype is not a documented Bedrock model ID.
  • No Anthropic Claude price is published from this offer file: prices are read only from curated usagetypes, and no curated entry is an Anthropic model. (When the collector was reviewed on 2026-10-29, the file listed only legacy Claude models, all without a service tier.)
  • Availability rests on AWS's own offer file listing an on-demand price for the model in us-east-1.

Kimi K3

Context window: 1M tokens

All rates, conditions and sources

Conditions of every rate below:

  • Listed for model "Kimi K3" (provider "Moonshot AI") in us-east-1 with service_tier "standard", as one USD price dimension from 0 to Inf.
  • AWS lists this price in USD per 1K tokens (unit "1K tokens"); it is multiplied by 1,000 exactly, by decimal arithmetic, to give USD per 1M tokens.
  • Only the standard on-demand tier in us-east-1 is published: flex, priority, batch and global cross-region tiers, other AWS Regions and provisioned throughput are excluded. The offer file lists cache prices for this model as separate usagetypes.

Input: $3.30 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-moonshotai.kimi-k3-mantle-input-tokens-standard.
  • Input tokens.

Output: $16.50 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-moonshotai.kimi-k3-mantle-output-tokens-standard.
  • Output tokens.

Cache read: $0.33 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-moonshotai.kimi-k3-mantle-cache-read-tokens-standard.
  • Cache read tokens (a "cache-read-tokens" usagetype). The offer file states no cache eligibility or cache lifetime for reads.

Cache write: $4.125 per 1M tokens · Standard tier · Region: us-east-1 · Cache lifetime: 30 minutes

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-moonshotai.kimi-k3-mantle-cache-write-tokens-30m-standard.
  • Cache write tokens with a 30-minute cache lifetime (a "cache-write-tokens-30m" usagetype), recorded as a 1800-second cache TTL.

Capability (Epoch AI): ECI 158.6 (90% range 156.6–160.8) · range spans bands · Epoch AI score for Kimi K3, model dated Jul 16, 2026

Price observed:

API availability: listed as available when checked (not a live service check)

Collection record

Primary source ↗ · provider documentation · Claims: bedrock-kimi-k3: specifications · Retrieved · Checked · Reviewed · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card ("Models at a glance") for Kimi K3, Model ID moonshotai.kimi-k3 (the offer's price list usagetype stem), raw SHA-256 cb7a08499bdfe48f069c28218e27714d00ea0f250adbc5d69531b56673c6e12f, read once on 2026-10-31 under the #837 AWS facts-plus-link basis: "Context window: 1M tokens".
  • Shorthand recorded as stated (#504); not converted to an exact token count. The value is AWS's statement for its Bedrock listing; another seller of the same model may state a different figure.
  • Specifications only: no pricing, availability, capability, lifecycle or release-date assertion.

Raw API evidence ↗ · provider API · Claims: bedrock-kimi-k3: pricing · bedrock-kimi-k3: availability · 122 more claims about 61 other catalog entries · Retrieved · Checked · Review due · Confidence: A

  • Raw SHA-256: f1ac6a175fb27b3ad80d2c5d8970be8c2e5f957906349e52d15ec11d43e97896; collector aws-bedrock-price-list-json-1.
  • AWS Price List Bulk API offer file for Amazon Bedrock (us-east-1): offer version 20261111013037, publicationDate 2026-11-11T01:30:37Z.
  • Reviewed on-demand token prices only: each curated model's exact usagetypes in us-east-1, with service_tier "standard" for standard-tier models and with no service_tier attribute for the reviewed untiered models, in US dollars as the offer file states them (hence confidence A). A hosted listing and pricing observation only: no checkpoint, developer, license, context-window, modality or release-date assertions, and the usagetype is not a documented Bedrock model ID.
  • No Anthropic Claude price is published from this offer file: prices are read only from curated usagetypes, and no curated entry is an Anthropic model. (When the collector was reviewed on 2026-10-29, the file listed only legacy Claude models, all without a service tier.)
  • Availability rests on AWS's own offer file listing an on-demand price for the model in us-east-1.

Llama 3 70B

Deprecated by provider · still listed · price checked Nov 12, 2026 (deprecation notice ↗)

All rates, conditions and sources

Conditions of every rate below:

  • Listed for model "Llama 3 70B" (provider "Meta") in us-east-1 with no service_tier attribute (feature "On-demand Inference"), as one USD price dimension from 0 to Inf.
  • AWS lists this price in USD per 1K tokens (unit "1K tokens"); it is multiplied by 1,000 exactly, by decimal arithmetic, to give USD per 1M tokens.
  • AWS's on-demand price for this model with no service tier stated: the row carries no service_tier attribute, so it is not labelled as the standard or any other tier. Only this plain on-demand usagetype in us-east-1 is published: any flex, priority, batch, latency-optimized, custom-model or cross-region (global) variant, other AWS Regions, provisioned throughput and model customization are excluded. The offer file lists no cache price for this model without a service tier; missing cache prices are unknown, not zero.

Input: $2.65 per 1M tokens · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-Llama3-70B-input-tokens.
  • Input tokens.

Output: $3.50 per 1M tokens · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-Llama3-70B-output-tokens.
  • Output tokens.

Price observed:

API availability: listed as available when checked (not a live service check)

Collection record

Primary source ↗ · provider documentation · Claims: bedrock-llama-3-70b: availability · Retrieved · Checked · Reviewed · Review due · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card for Llama 3 70B Instruct, Model ID meta.llama3-70b-instruct-v1:0, raw SHA-256 313758eff6b2cc826329be3f2daf61f427e7fad35cc1fb4811ea1f8f6b3970f2, read once at 2026-11-02T11:39:03Z under the #837 AWS facts-plus-link basis: "Model lifecycle: Legacy"; "Model EOL date: March 30, 2027"; "Model lifecycle policy: Model lifecycle (For Models Launched Prior to Sept 7 2026)".
  • Amazon Bedrock "Model lifecycle (Legacy)" page (https://docs.aws.amazon.com/bedrock/latest/userguide/model-lifecycle-legacy.html), read once at 2026-11-02T11:37:59Z, raw SHA-256 c33d9d25d7da7f47c098aa864830361de770224f38a790070bada021dac56850: its Legacy table lists Llama 3 70B Instruct (meta.llama3-70b-instruct-v1:0) with Legacy date September 30, 2026, EOL date March 30, 2027 and public extended access start date "To be announced". The page states: "New customers can't use Legacy models and existing customers may lose access to Legacy models after 15 days of inactivity."
  • That table row's Regions (ap-south-1, ca-central-1, eu-west-2, us-east-1, us-west-2) include us-east-1, the region this offer is published for.
  • Availability stays available: the us-east-1 price list still lists an on-demand price (2026-11-06 archive), and the page says "you can continue to use a Legacy model during this period" (the Legacy period before the EOL date), though not as a new customer. The EOL date is in the future, so no retirement is recorded; reviewAfter is the EOL date, a re-review date, not a retirement.
  • Reviewed by Claude (owner delegation 2026-09-27), #863.

Raw API evidence ↗ · provider API · Claims: bedrock-llama-3-70b: pricing · bedrock-llama-3-70b: availability · 122 more claims about 61 other catalog entries · Retrieved · Checked · Review due · Confidence: A

  • Raw SHA-256: f1ac6a175fb27b3ad80d2c5d8970be8c2e5f957906349e52d15ec11d43e97896; collector aws-bedrock-price-list-json-1.
  • AWS Price List Bulk API offer file for Amazon Bedrock (us-east-1): offer version 20261111013037, publicationDate 2026-11-11T01:30:37Z.
  • Reviewed on-demand token prices only: each curated model's exact usagetypes in us-east-1, with service_tier "standard" for standard-tier models and with no service_tier attribute for the reviewed untiered models, in US dollars as the offer file states them (hence confidence A). A hosted listing and pricing observation only: no checkpoint, developer, license, context-window, modality or release-date assertions, and the usagetype is not a documented Bedrock model ID.
  • No Anthropic Claude price is published from this offer file: prices are read only from curated usagetypes, and no curated entry is an Anthropic model. (When the collector was reviewed on 2026-10-29, the file listed only legacy Claude models, all without a service tier.)
  • Availability rests on AWS's own offer file listing an on-demand price for the model in us-east-1.

Llama 3 8B

Deprecated by provider · still listed · price checked Nov 12, 2026 (deprecation notice ↗)

All rates, conditions and sources

Conditions of every rate below:

  • Listed for model "Llama 3 8B" (provider "Meta") in us-east-1 with no service_tier attribute (feature "On-demand Inference"), as one USD price dimension from 0 to Inf.
  • AWS lists this price in USD per 1K tokens (unit "1K tokens"); it is multiplied by 1,000 exactly, by decimal arithmetic, to give USD per 1M tokens.
  • AWS's on-demand price for this model with no service tier stated: the row carries no service_tier attribute, so it is not labelled as the standard or any other tier. Only this plain on-demand usagetype in us-east-1 is published: any flex, priority, batch, latency-optimized, custom-model or cross-region (global) variant, other AWS Regions, provisioned throughput and model customization are excluded. The offer file lists no cache price for this model without a service tier; missing cache prices are unknown, not zero.

Input: $0.30 per 1M tokens · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-Llama3-8B-input-tokens.
  • Input tokens.

Output: $0.60 per 1M tokens · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-Llama3-8B-output-tokens.
  • Output tokens.

Price observed:

API availability: listed as available when checked (not a live service check)

Collection record

Primary source ↗ · provider documentation · Claims: bedrock-llama-3-8b: availability · Retrieved · Checked · Reviewed · Review due · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card for Llama 3 8B Instruct, Model ID meta.llama3-8b-instruct-v1:0, raw SHA-256 23e1e916414f7eb50301fe09ecbb315d460ab3b76c28cc60dd6ca7fbc05791d0, read once at 2026-11-02T11:38:56Z under the #837 AWS facts-plus-link basis: "Model lifecycle: Legacy"; "Model EOL date: March 30, 2027"; "Model lifecycle policy: Model lifecycle (For Models Launched Prior to Sept 7 2026)".
  • Amazon Bedrock "Model lifecycle (Legacy)" page (https://docs.aws.amazon.com/bedrock/latest/userguide/model-lifecycle-legacy.html), read once at 2026-11-02T11:37:59Z, raw SHA-256 c33d9d25d7da7f47c098aa864830361de770224f38a790070bada021dac56850: its Legacy table lists Llama 3 8B Instruct (meta.llama3-8b-instruct-v1:0) with Legacy date September 30, 2026, EOL date March 30, 2027 and public extended access start date "To be announced". The page states: "New customers can't use Legacy models and existing customers may lose access to Legacy models after 15 days of inactivity."
  • That table row's Regions (ap-south-1, ca-central-1, eu-west-2, us-east-1, us-west-2) include us-east-1, the region this offer is published for.
  • Availability stays available: the us-east-1 price list still lists an on-demand price (2026-11-06 archive), and the page says "you can continue to use a Legacy model during this period" (the Legacy period before the EOL date), though not as a new customer. The EOL date is in the future, so no retirement is recorded; reviewAfter is the EOL date, a re-review date, not a retirement.
  • Reviewed by Claude (owner delegation 2026-09-27), #863.

Raw API evidence ↗ · provider API · Claims: bedrock-llama-3-8b: pricing · bedrock-llama-3-8b: availability · 122 more claims about 61 other catalog entries · Retrieved · Checked · Review due · Confidence: A

  • Raw SHA-256: f1ac6a175fb27b3ad80d2c5d8970be8c2e5f957906349e52d15ec11d43e97896; collector aws-bedrock-price-list-json-1.
  • AWS Price List Bulk API offer file for Amazon Bedrock (us-east-1): offer version 20261111013037, publicationDate 2026-11-11T01:30:37Z.
  • Reviewed on-demand token prices only: each curated model's exact usagetypes in us-east-1, with service_tier "standard" for standard-tier models and with no service_tier attribute for the reviewed untiered models, in US dollars as the offer file states them (hence confidence A). A hosted listing and pricing observation only: no checkpoint, developer, license, context-window, modality or release-date assertions, and the usagetype is not a documented Bedrock model ID.
  • No Anthropic Claude price is published from this offer file: prices are read only from curated usagetypes, and no curated entry is an Anthropic model. (When the collector was reviewed on 2026-10-29, the file listed only legacy Claude models, all without a service tier.)
  • Availability rests on AWS's own offer file listing an on-demand price for the model in us-east-1.

Llama 3.1 70B

Deprecated by provider · still listed · price checked Nov 12, 2026 (deprecation notice ↗)

All rates, conditions and sources

Conditions of every rate below:

  • Listed for model "Llama 3.1 70B" (provider "Meta") in us-east-1 with no service_tier attribute (feature "On-demand Inference"), as one USD price dimension from 0 to Inf.
  • AWS lists this price in USD per 1K tokens (unit "1K tokens"); it is multiplied by 1,000 exactly, by decimal arithmetic, to give USD per 1M tokens.
  • AWS's on-demand price for this model with no service tier stated: the row carries no service_tier attribute, so it is not labelled as the standard or any other tier. Only this plain on-demand usagetype in us-east-1 is published: any flex, priority, batch, latency-optimized, custom-model or cross-region (global) variant, other AWS Regions, provisioned throughput and model customization are excluded. The offer file lists no cache price for this model without a service tier; missing cache prices are unknown, not zero.

Input: $0.72 per 1M tokens · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-Llama3-1-70B-input-tokens.
  • Input tokens.

Output: $0.72 per 1M tokens · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-Llama3-1-70B-output-tokens.
  • Output tokens.

Price observed:

API availability: listed as available when checked (not a live service check)

Collection record

Primary source ↗ · provider documentation · Claims: bedrock-llama-3-1-70b: availability · Retrieved · Checked · Reviewed · Review due · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card for Llama 3.1 70B Instruct, Model ID meta.llama3-1-70b-instruct-v1:0, raw SHA-256 aadc72f76f4ed80cb4624074b61f0bb8a4e871c3fa62382218b861dbf14e680a, read once at 2026-11-02T11:39:15Z under the #837 AWS facts-plus-link basis: "Model lifecycle: Legacy"; "Model EOL date: March 30, 2027"; "Model lifecycle policy: Model lifecycle (For Models Launched Prior to Sept 7 2026)".
  • Amazon Bedrock "Model lifecycle (Legacy)" page (https://docs.aws.amazon.com/bedrock/latest/userguide/model-lifecycle-legacy.html), read once at 2026-11-02T11:37:59Z, raw SHA-256 c33d9d25d7da7f47c098aa864830361de770224f38a790070bada021dac56850: its Legacy table lists Llama 3.1 70B Instruct (meta.llama3-1-70b-instruct-v1:0) with Legacy date September 30, 2026, EOL date March 30, 2027 and public extended access start date "To be announced". The page states: "New customers can't use Legacy models and existing customers may lose access to Legacy models after 15 days of inactivity."
  • That table row's Regions (us-east-1, us-east-2, us-west-2) include us-east-1, the region this offer is published for.
  • Availability stays available: the us-east-1 price list still lists an on-demand price (2026-11-06 archive), and the page says "you can continue to use a Legacy model during this period" (the Legacy period before the EOL date), though not as a new customer. The EOL date is in the future, so no retirement is recorded; reviewAfter is the EOL date, a re-review date, not a retirement.
  • Reviewed by Claude (owner delegation 2026-09-27), #863.

Raw API evidence ↗ · provider API · Claims: bedrock-llama-3-1-70b: pricing · bedrock-llama-3-1-70b: availability · 122 more claims about 61 other catalog entries · Retrieved · Checked · Review due · Confidence: A

  • Raw SHA-256: f1ac6a175fb27b3ad80d2c5d8970be8c2e5f957906349e52d15ec11d43e97896; collector aws-bedrock-price-list-json-1.
  • AWS Price List Bulk API offer file for Amazon Bedrock (us-east-1): offer version 20261111013037, publicationDate 2026-11-11T01:30:37Z.
  • Reviewed on-demand token prices only: each curated model's exact usagetypes in us-east-1, with service_tier "standard" for standard-tier models and with no service_tier attribute for the reviewed untiered models, in US dollars as the offer file states them (hence confidence A). A hosted listing and pricing observation only: no checkpoint, developer, license, context-window, modality or release-date assertions, and the usagetype is not a documented Bedrock model ID.
  • No Anthropic Claude price is published from this offer file: prices are read only from curated usagetypes, and no curated entry is an Anthropic model. (When the collector was reviewed on 2026-10-29, the file listed only legacy Claude models, all without a service tier.)
  • Availability rests on AWS's own offer file listing an on-demand price for the model in us-east-1.

Llama 3.1 8B

Deprecated by provider · still listed · price checked Nov 12, 2026 (deprecation notice ↗)

All rates, conditions and sources

Conditions of every rate below:

  • Listed for model "Llama 3.1 8B" (provider "Meta") in us-east-1 with no service_tier attribute (feature "On-demand Inference"), as one USD price dimension from 0 to Inf.
  • AWS lists this price in USD per 1K tokens (unit "1K tokens"); it is multiplied by 1,000 exactly, by decimal arithmetic, to give USD per 1M tokens.
  • AWS's on-demand price for this model with no service tier stated: the row carries no service_tier attribute, so it is not labelled as the standard or any other tier. Only this plain on-demand usagetype in us-east-1 is published: any flex, priority, batch, latency-optimized, custom-model or cross-region (global) variant, other AWS Regions, provisioned throughput and model customization are excluded. The offer file lists no cache price for this model without a service tier; missing cache prices are unknown, not zero.

Input: $0.22 per 1M tokens · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-Llama3-1-8B-input-tokens.
  • Input tokens.

Output: $0.22 per 1M tokens · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-Llama3-1-8B-output-tokens.
  • Output tokens.

Price observed:

API availability: listed as available when checked (not a live service check)

Collection record

Primary source ↗ · provider documentation · Claims: bedrock-llama-3-1-8b: availability · Retrieved · Checked · Reviewed · Review due · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card for Llama 3.1 8B Instruct, Model ID meta.llama3-1-8b-instruct-v1:0, raw SHA-256 49ea598aa3e7f95c7351eb2d02ed266e32425e868d03413a8ce30fde59f35b47, read once at 2026-11-02T11:39:09Z under the #837 AWS facts-plus-link basis: "Model lifecycle: Legacy"; "Model EOL date: March 30, 2027"; "Model lifecycle policy: Model lifecycle (For Models Launched Prior to Sept 7 2026)".
  • Amazon Bedrock "Model lifecycle (Legacy)" page (https://docs.aws.amazon.com/bedrock/latest/userguide/model-lifecycle-legacy.html), read once at 2026-11-02T11:37:59Z, raw SHA-256 c33d9d25d7da7f47c098aa864830361de770224f38a790070bada021dac56850: its Legacy table lists Llama 3.1 8B Instruct (meta.llama3-1-8b-instruct-v1:0) with Legacy date September 30, 2026, EOL date March 30, 2027 and public extended access start date "To be announced". The page states: "New customers can't use Legacy models and existing customers may lose access to Legacy models after 15 days of inactivity."
  • That table row's Regions (us-east-1, us-east-2, us-west-2) include us-east-1, the region this offer is published for.
  • Availability stays available: the us-east-1 price list still lists an on-demand price (2026-11-06 archive), and the page says "you can continue to use a Legacy model during this period" (the Legacy period before the EOL date), though not as a new customer. The EOL date is in the future, so no retirement is recorded; reviewAfter is the EOL date, a re-review date, not a retirement.
  • Reviewed by Claude (owner delegation 2026-09-27), #863.

Raw API evidence ↗ · provider API · Claims: bedrock-llama-3-1-8b: pricing · bedrock-llama-3-1-8b: availability · 122 more claims about 61 other catalog entries · Retrieved · Checked · Review due · Confidence: A

  • Raw SHA-256: f1ac6a175fb27b3ad80d2c5d8970be8c2e5f957906349e52d15ec11d43e97896; collector aws-bedrock-price-list-json-1.
  • AWS Price List Bulk API offer file for Amazon Bedrock (us-east-1): offer version 20261111013037, publicationDate 2026-11-11T01:30:37Z.
  • Reviewed on-demand token prices only: each curated model's exact usagetypes in us-east-1, with service_tier "standard" for standard-tier models and with no service_tier attribute for the reviewed untiered models, in US dollars as the offer file states them (hence confidence A). A hosted listing and pricing observation only: no checkpoint, developer, license, context-window, modality or release-date assertions, and the usagetype is not a documented Bedrock model ID.
  • No Anthropic Claude price is published from this offer file: prices are read only from curated usagetypes, and no curated entry is an Anthropic model. (When the collector was reviewed on 2026-10-29, the file listed only legacy Claude models, all without a service tier.)
  • Availability rests on AWS's own offer file listing an on-demand price for the model in us-east-1.

Llama 3.3 70B

Deprecated by provider · still listed · price checked Nov 12, 2026 (deprecation notice ↗)

All rates, conditions and sources

Conditions of every rate below:

  • Listed for model "Llama 3.3 70B" (provider "Meta") in us-east-1 with no service_tier attribute (feature "On-demand Inference"), as one USD price dimension from 0 to Inf.
  • AWS lists this price in USD per 1K tokens (unit "1K tokens"); it is multiplied by 1,000 exactly, by decimal arithmetic, to give USD per 1M tokens.
  • AWS's on-demand price for this model with no service tier stated: the row carries no service_tier attribute, so it is not labelled as the standard or any other tier. Only this plain on-demand usagetype in us-east-1 is published: any flex, priority, batch, latency-optimized, custom-model or cross-region (global) variant, other AWS Regions, provisioned throughput and model customization are excluded. The offer file lists no cache price for this model without a service tier; missing cache prices are unknown, not zero.

Input: $0.72 per 1M tokens · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-Llama3-3-70B-input-tokens.
  • Input tokens.

Output: $0.72 per 1M tokens · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-Llama3-3-70B-output-tokens.
  • Output tokens.

Price observed:

API availability: listed as available when checked (not a live service check)

Collection record

Primary source ↗ · provider documentation · Claims: bedrock-llama-3-3-70b: availability · Retrieved · Checked · Reviewed · Review due · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card for Llama 3.3 70B Instruct, Model ID meta.llama3-3-70b-instruct-v1:0, raw SHA-256 6eb8e11f84851e760de14e5196d01c5643ee5c4b6de911da3e0882d5e0fa7704, read once at 2026-11-02T11:39:46Z under the #837 AWS facts-plus-link basis: "Model lifecycle: Legacy"; "Model EOL date: March 30, 2027"; "Model lifecycle policy: Model lifecycle (For Models Launched Prior to Sept 7 2026)".
  • Amazon Bedrock "Model lifecycle (Legacy)" page (https://docs.aws.amazon.com/bedrock/latest/userguide/model-lifecycle-legacy.html), read once at 2026-11-02T11:37:59Z, raw SHA-256 c33d9d25d7da7f47c098aa864830361de770224f38a790070bada021dac56850: its Legacy table lists Llama 3.3 70B Instruct (meta.llama3-3-70b-instruct-v1:0) with Legacy date September 30, 2026, EOL date March 30, 2027 and public extended access start date "To be announced". The page states: "New customers can't use Legacy models and existing customers may lose access to Legacy models after 15 days of inactivity."
  • That table row's Regions (us-east-1, us-east-2, us-west-2) include us-east-1, the region this offer is published for.
  • Availability stays available: the us-east-1 price list still lists an on-demand price (2026-11-06 archive), and the page says "you can continue to use a Legacy model during this period" (the Legacy period before the EOL date), though not as a new customer. The EOL date is in the future, so no retirement is recorded; reviewAfter is the EOL date, a re-review date, not a retirement.
  • Reviewed by Claude (owner delegation 2026-09-27), #863.

Raw API evidence ↗ · provider API · Claims: bedrock-llama-3-3-70b: pricing · bedrock-llama-3-3-70b: availability · 122 more claims about 61 other catalog entries · Retrieved · Checked · Review due · Confidence: A

  • Raw SHA-256: f1ac6a175fb27b3ad80d2c5d8970be8c2e5f957906349e52d15ec11d43e97896; collector aws-bedrock-price-list-json-1.
  • AWS Price List Bulk API offer file for Amazon Bedrock (us-east-1): offer version 20261111013037, publicationDate 2026-11-11T01:30:37Z.
  • Reviewed on-demand token prices only: each curated model's exact usagetypes in us-east-1, with service_tier "standard" for standard-tier models and with no service_tier attribute for the reviewed untiered models, in US dollars as the offer file states them (hence confidence A). A hosted listing and pricing observation only: no checkpoint, developer, license, context-window, modality or release-date assertions, and the usagetype is not a documented Bedrock model ID.
  • No Anthropic Claude price is published from this offer file: prices are read only from curated usagetypes, and no curated entry is an Anthropic model. (When the collector was reviewed on 2026-10-29, the file listed only legacy Claude models, all without a service tier.)
  • Availability rests on AWS's own offer file listing an on-demand price for the model in us-east-1.

Llama 4 Maverick 17B

Deprecated by provider · still listed · price checked Nov 12, 2026 (deprecation notice ↗)

All rates, conditions and sources

Conditions of every rate below:

  • Listed for model "Llama 4 Maverick 17B" (provider "Meta") in us-east-1 with no service_tier attribute (feature "On-demand Inference"), as one USD price dimension from 0 to Inf.
  • AWS lists this price in USD per 1K tokens (unit "1K tokens"); it is multiplied by 1,000 exactly, by decimal arithmetic, to give USD per 1M tokens.
  • AWS's on-demand price for this model with no service tier stated: the row carries no service_tier attribute, so it is not labelled as the standard or any other tier. Only this plain on-demand usagetype in us-east-1 is published: any flex, priority, batch, latency-optimized, custom-model or cross-region (global) variant, other AWS Regions, provisioned throughput and model customization are excluded. The offer file lists no cache price for this model without a service tier; missing cache prices are unknown, not zero.

Input: $0.24 per 1M tokens · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-Llama4-Maverick-17B-input-tokens.
  • Input tokens.

Output: $0.97 per 1M tokens · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-Llama4-Maverick-17B-output-tokens.
  • Output tokens.

Price observed:

API availability: listed as available when checked (not a live service check)

Collection record

Primary source ↗ · provider documentation · Claims: bedrock-llama-4-maverick-17b: availability · Retrieved · Checked · Reviewed · Review due · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card for Llama 4 Maverick 17B Instruct, Model ID meta.llama4-maverick-17b-instruct-v1:0, raw SHA-256 c412804e07c192a128cf77c886372144bfecb7708ea921e92c7510e5b0456848, read once at 2026-11-02T11:39:58Z under the #837 AWS facts-plus-link basis: "Model lifecycle: Legacy"; "Model EOL date: March 30, 2027"; "Model lifecycle policy: Model lifecycle (For Models Launched Prior to Sept 7 2026)".
  • Amazon Bedrock "Model lifecycle (Legacy)" page (https://docs.aws.amazon.com/bedrock/latest/userguide/model-lifecycle-legacy.html), read once at 2026-11-02T11:37:59Z, raw SHA-256 c33d9d25d7da7f47c098aa864830361de770224f38a790070bada021dac56850: its Legacy table lists Llama 4 Maverick 17B Instruct (meta.llama4-maverick-17b-instruct-v1:0) with Legacy date September 30, 2026, EOL date March 30, 2027 and public extended access start date "To be announced". The page states: "New customers can't use Legacy models and existing customers may lose access to Legacy models after 15 days of inactivity."
  • That table row's Regions (us-east-1, us-east-2, us-west-1, us-west-2) include us-east-1, the region this offer is published for.
  • Availability stays available: the us-east-1 price list still lists an on-demand price (2026-11-06 archive), and the page says "you can continue to use a Legacy model during this period" (the Legacy period before the EOL date), though not as a new customer. The EOL date is in the future, so no retirement is recorded; reviewAfter is the EOL date, a re-review date, not a retirement.
  • Reviewed by Claude (owner delegation 2026-09-27), #863.

Raw API evidence ↗ · provider API · Claims: bedrock-llama-4-maverick-17b: pricing · bedrock-llama-4-maverick-17b: availability · 122 more claims about 61 other catalog entries · Retrieved · Checked · Review due · Confidence: A

  • Raw SHA-256: f1ac6a175fb27b3ad80d2c5d8970be8c2e5f957906349e52d15ec11d43e97896; collector aws-bedrock-price-list-json-1.
  • AWS Price List Bulk API offer file for Amazon Bedrock (us-east-1): offer version 20261111013037, publicationDate 2026-11-11T01:30:37Z.
  • Reviewed on-demand token prices only: each curated model's exact usagetypes in us-east-1, with service_tier "standard" for standard-tier models and with no service_tier attribute for the reviewed untiered models, in US dollars as the offer file states them (hence confidence A). A hosted listing and pricing observation only: no checkpoint, developer, license, context-window, modality or release-date assertions, and the usagetype is not a documented Bedrock model ID.
  • No Anthropic Claude price is published from this offer file: prices are read only from curated usagetypes, and no curated entry is an Anthropic model. (When the collector was reviewed on 2026-10-29, the file listed only legacy Claude models, all without a service tier.)
  • Availability rests on AWS's own offer file listing an on-demand price for the model in us-east-1.

Llama 4 Scout 17B

Deprecated by provider · still listed · price checked Nov 12, 2026 (deprecation notice ↗)

All rates, conditions and sources

Conditions of every rate below:

  • Listed for model "Llama 4 Scout 17B" (provider "Meta") in us-east-1 with no service_tier attribute (feature "On-demand Inference"), as one USD price dimension from 0 to Inf.
  • AWS lists this price in USD per 1K tokens (unit "1K tokens"); it is multiplied by 1,000 exactly, by decimal arithmetic, to give USD per 1M tokens.
  • AWS's on-demand price for this model with no service tier stated: the row carries no service_tier attribute, so it is not labelled as the standard or any other tier. Only this plain on-demand usagetype in us-east-1 is published: any flex, priority, batch, latency-optimized, custom-model or cross-region (global) variant, other AWS Regions, provisioned throughput and model customization are excluded. The offer file lists no cache price for this model without a service tier; missing cache prices are unknown, not zero.

Input: $0.17 per 1M tokens · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-Llama4-Scout-17B-input-tokens.
  • Input tokens.

Output: $0.66 per 1M tokens · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-Llama4-Scout-17B-output-tokens.
  • Output tokens.

Price observed:

API availability: listed as available when checked (not a live service check)

Collection record

Primary source ↗ · provider documentation · Claims: bedrock-llama-4-scout-17b: availability · Retrieved · Checked · Reviewed · Review due · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card for Llama 4 Scout 17B Instruct, Model ID meta.llama4-scout-17b-instruct-v1:0, raw SHA-256 4f55043756cfde53254e13cea19e5fc81238b86eda24510681c11f250497151f, read once at 2026-11-02T11:39:52Z under the #837 AWS facts-plus-link basis: "Model lifecycle: Legacy"; "Model EOL date: March 30, 2027"; "Model lifecycle policy: Model lifecycle (For Models Launched Prior to Sept 7 2026)".
  • Amazon Bedrock "Model lifecycle (Legacy)" page (https://docs.aws.amazon.com/bedrock/latest/userguide/model-lifecycle-legacy.html), read once at 2026-11-02T11:37:59Z, raw SHA-256 c33d9d25d7da7f47c098aa864830361de770224f38a790070bada021dac56850: its Legacy table lists Llama 4 Scout 17B Instruct (meta.llama4-scout-17b-instruct-v1:0) with Legacy date September 30, 2026, EOL date March 30, 2027 and public extended access start date "To be announced". The page states: "New customers can't use Legacy models and existing customers may lose access to Legacy models after 15 days of inactivity."
  • That table row's Regions (us-east-1, us-east-2, us-west-1, us-west-2) include us-east-1, the region this offer is published for.
  • Availability stays available: the us-east-1 price list still lists an on-demand price (2026-11-06 archive), and the page says "you can continue to use a Legacy model during this period" (the Legacy period before the EOL date), though not as a new customer. The EOL date is in the future, so no retirement is recorded; reviewAfter is the EOL date, a re-review date, not a retirement.
  • Reviewed by Claude (owner delegation 2026-09-27), #863.

Raw API evidence ↗ · provider API · Claims: bedrock-llama-4-scout-17b: pricing · bedrock-llama-4-scout-17b: availability · 122 more claims about 61 other catalog entries · Retrieved · Checked · Review due · Confidence: A

  • Raw SHA-256: f1ac6a175fb27b3ad80d2c5d8970be8c2e5f957906349e52d15ec11d43e97896; collector aws-bedrock-price-list-json-1.
  • AWS Price List Bulk API offer file for Amazon Bedrock (us-east-1): offer version 20261111013037, publicationDate 2026-11-11T01:30:37Z.
  • Reviewed on-demand token prices only: each curated model's exact usagetypes in us-east-1, with service_tier "standard" for standard-tier models and with no service_tier attribute for the reviewed untiered models, in US dollars as the offer file states them (hence confidence A). A hosted listing and pricing observation only: no checkpoint, developer, license, context-window, modality or release-date assertions, and the usagetype is not a documented Bedrock model ID.
  • No Anthropic Claude price is published from this offer file: prices are read only from curated usagetypes, and no curated entry is an Anthropic model. (When the collector was reviewed on 2026-10-29, the file listed only legacy Claude models, all without a service tier.)
  • Availability rests on AWS's own offer file listing an on-demand price for the model in us-east-1.

Magistral Small 1.2

Context window: 128K tokens

All rates, conditions and sources

Conditions of every rate below:

  • Listed for model "Magistral Small 1.2" (provider "Mistral") in us-east-1 with service_tier "standard", as one USD price dimension from 0 to Inf.
  • AWS lists this price in USD per 1K tokens (unit "1K tokens"); it is multiplied by 1,000 exactly, by decimal arithmetic, to give USD per 1M tokens.
  • Only the standard on-demand tier in us-east-1 is published: flex, priority, batch and global cross-region tiers, other AWS Regions and provisioned throughput are excluded. The offer file lists no standard-tier cache price for this model; missing cache prices are unknown, not zero.

Input: $0.50 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-mistral.magistral-small-2509-mantle-input-tokens-standard.
  • Input tokens.

Output: $1.50 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-mistral.magistral-small-2509-mantle-output-tokens-standard.
  • Output tokens.

Price observed:

API availability: listed as available when checked (not a live service check)

Collection record

Primary source ↗ · provider documentation · Claims: bedrock-magistral-small-2509: specifications · Retrieved · Checked · Reviewed · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card ("Models at a glance") for Magistral Small 2509, Model ID mistral.magistral-small-2509 (the offer's price list usagetype stem), raw SHA-256 3e36089edadf84be603222ce0d572725253b055509b15f5ef4529693fc01b4b5, read once on 2026-10-31 under the #837 AWS facts-plus-link basis: "Context window: 128K tokens".
  • Shorthand recorded as stated (#504); not converted to an exact token count. The value is AWS's statement for its Bedrock listing; another seller of the same model may state a different figure.
  • Specifications only: no pricing, availability, capability, lifecycle or release-date assertion.

Raw API evidence ↗ · provider API · Claims: bedrock-magistral-small-2509: pricing · bedrock-magistral-small-2509: availability · 122 more claims about 61 other catalog entries · Retrieved · Checked · Review due · Confidence: A

  • Raw SHA-256: f1ac6a175fb27b3ad80d2c5d8970be8c2e5f957906349e52d15ec11d43e97896; collector aws-bedrock-price-list-json-1.
  • AWS Price List Bulk API offer file for Amazon Bedrock (us-east-1): offer version 20261111013037, publicationDate 2026-11-11T01:30:37Z.
  • Reviewed on-demand token prices only: each curated model's exact usagetypes in us-east-1, with service_tier "standard" for standard-tier models and with no service_tier attribute for the reviewed untiered models, in US dollars as the offer file states them (hence confidence A). A hosted listing and pricing observation only: no checkpoint, developer, license, context-window, modality or release-date assertions, and the usagetype is not a documented Bedrock model ID.
  • No Anthropic Claude price is published from this offer file: prices are read only from curated usagetypes, and no curated entry is an Anthropic model. (When the collector was reviewed on 2026-10-29, the file listed only legacy Claude models, all without a service tier.)
  • Availability rests on AWS's own offer file listing an on-demand price for the model in us-east-1.

MiniMax M2

Context window: 1M tokens

Deprecated by provider · still listed · price checked Nov 12, 2026 (deprecation notice ↗)

All rates, conditions and sources

Conditions of every rate below:

  • Listed for model "Minimax M2" (provider "Minimax AI") in us-east-1 with service_tier "standard", as one USD price dimension from 0 to Inf.
  • AWS lists this price in USD per 1K tokens (unit "1K tokens"); it is multiplied by 1,000 exactly, by decimal arithmetic, to give USD per 1M tokens.
  • Only the standard on-demand tier in us-east-1 is published: flex, priority, batch and global cross-region tiers, other AWS Regions and provisioned throughput are excluded. The offer file lists no standard-tier cache price for this model; missing cache prices are unknown, not zero.

Input: $0.30 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-minimax.minimax-m2-mantle-input-tokens-standard.
  • Input tokens.

Output: $1.20 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-minimax.minimax-m2-mantle-output-tokens-standard.
  • Output tokens.

Price observed:

API availability: listed as available when checked (not a live service check)

Collection record

Primary source ↗ · provider documentation · Claims: bedrock-minimax-m2: specifications · Retrieved · Checked · Reviewed · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card ("Models at a glance") for MiniMax M2, Model ID minimax.minimax-m2 (the offer's price list usagetype stem), raw SHA-256 058a1ce9dad3a04cbac59d7f6a3277ae9246a628e211143ebcc375f41534b09e, read once on 2026-10-31 under the #837 AWS facts-plus-link basis: "Context window: 1M tokens".
  • Shorthand recorded as stated (#504); not converted to an exact token count. The value is AWS's statement for its Bedrock listing; another seller of the same model may state a different figure.
  • Specifications only: no pricing, availability, capability, lifecycle or release-date assertion.

Primary source ↗ · provider documentation · Claims: bedrock-minimax-m2: availability · Retrieved · Checked · Reviewed · Review due · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card for MiniMax M2, Model ID minimax.minimax-m2, raw SHA-256 058a1ce9dad3a04cbac59d7f6a3277ae9246a628e211143ebcc375f41534b09e, read once on 2026-10-31 under the #837 AWS facts-plus-link basis: "Model lifecycle: Legacy"; "Model EOL date: March 30, 2027"; "Model lifecycle policy: Model lifecycle (For Models Launched Prior to Sept 7 2026)".
  • Amazon Bedrock "Model lifecycle (Legacy)" page (https://docs.aws.amazon.com/bedrock/latest/userguide/model-lifecycle-legacy.html), read once at 2026-10-31T16:36:03Z, raw SHA-256 c33d9d25d7da7f47c098aa864830361de770224f38a790070bada021dac56850: its Legacy table lists MiniMax M2 (minimax.minimax-m2) with Legacy date September 30, 2026, EOL date March 30, 2027 and public extended access start date "To be announced". The page states: "New customers can't use Legacy models and existing customers may lose access to Legacy models after 15 days of inactivity."
  • That table row's Regions include us-east-1, the region this offer is published for.
  • Availability stays available: the us-east-1 price list still lists an on-demand price, and the page says "you can continue to use a Legacy model during this period" (the Legacy period before the EOL date), though not as a new customer. The EOL date is in the future, so no retirement is recorded; reviewAfter is the EOL date, a re-review date, not a retirement.
  • Reviewed by Claude (owner delegation 2026-09-27), #909.

Raw API evidence ↗ · provider API · Claims: bedrock-minimax-m2: pricing · bedrock-minimax-m2: availability · 122 more claims about 61 other catalog entries · Retrieved · Checked · Review due · Confidence: A

  • Raw SHA-256: f1ac6a175fb27b3ad80d2c5d8970be8c2e5f957906349e52d15ec11d43e97896; collector aws-bedrock-price-list-json-1.
  • AWS Price List Bulk API offer file for Amazon Bedrock (us-east-1): offer version 20261111013037, publicationDate 2026-11-11T01:30:37Z.
  • Reviewed on-demand token prices only: each curated model's exact usagetypes in us-east-1, with service_tier "standard" for standard-tier models and with no service_tier attribute for the reviewed untiered models, in US dollars as the offer file states them (hence confidence A). A hosted listing and pricing observation only: no checkpoint, developer, license, context-window, modality or release-date assertions, and the usagetype is not a documented Bedrock model ID.
  • No Anthropic Claude price is published from this offer file: prices are read only from curated usagetypes, and no curated entry is an Anthropic model. (When the collector was reviewed on 2026-10-29, the file listed only legacy Claude models, all without a service tier.)
  • Availability rests on AWS's own offer file listing an on-demand price for the model in us-east-1.

MiniMax M2.1

Context window: 196K tokens

Deprecated by provider · still listed · price checked Nov 12, 2026 (deprecation notice ↗)

All rates, conditions and sources

Conditions of every rate below:

  • Listed for model "Minimax M2.1" (provider "Minimax AI") in us-east-1 with service_tier "standard", as one USD price dimension from 0 to Inf.
  • AWS lists this price in USD per 1K tokens (unit "1K tokens"); it is multiplied by 1,000 exactly, by decimal arithmetic, to give USD per 1M tokens.
  • Only the standard on-demand tier in us-east-1 is published: flex, priority, batch and global cross-region tiers, other AWS Regions and provisioned throughput are excluded. The offer file lists no standard-tier cache price for this model; missing cache prices are unknown, not zero.

Input: $0.30 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-minimax.minimax-m2.1-mantle-input-tokens-standard.
  • Input tokens.

Output: $1.20 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-minimax.minimax-m2.1-mantle-output-tokens-standard.
  • Output tokens.

Price observed:

API availability: listed as available when checked (not a live service check)

Collection record

Primary source ↗ · provider documentation · Claims: bedrock-minimax-m2-1: specifications · Retrieved · Checked · Reviewed · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card ("Models at a glance") for MiniMax M2.1, Model ID minimax.minimax-m2.1 (the offer's price list usagetype stem), raw SHA-256 9888437d54ec6c6064613cf4aeba5dabc6bb238e500697b856d5b759dd7b430a, read once on 2026-10-31 under the #837 AWS facts-plus-link basis: "Context window: 196K tokens".
  • Shorthand recorded as stated (#504); not converted to an exact token count. The value is AWS's statement for its Bedrock listing; another seller of the same model may state a different figure.
  • Specifications only: no pricing, availability, capability, lifecycle or release-date assertion.

Primary source ↗ · provider documentation · Claims: bedrock-minimax-m2-1: availability · Retrieved · Checked · Reviewed · Review due · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card for MiniMax M2.1, Model ID minimax.minimax-m2.1, raw SHA-256 9888437d54ec6c6064613cf4aeba5dabc6bb238e500697b856d5b759dd7b430a, read once on 2026-10-31 under the #837 AWS facts-plus-link basis: "Model lifecycle: Legacy"; "Model EOL date: March 30, 2027"; "Model lifecycle policy: Model lifecycle (For Models Launched Prior to Sept 7 2026)".
  • Amazon Bedrock "Model lifecycle (Legacy)" page (https://docs.aws.amazon.com/bedrock/latest/userguide/model-lifecycle-legacy.html), read once at 2026-10-31T16:36:03Z, raw SHA-256 c33d9d25d7da7f47c098aa864830361de770224f38a790070bada021dac56850: its Legacy table lists MiniMax M2.1 (minimax.minimax-m2.1) with Legacy date September 30, 2026, EOL date March 30, 2027 and public extended access start date "To be announced". The page states: "New customers can't use Legacy models and existing customers may lose access to Legacy models after 15 days of inactivity."
  • That table row's Regions include us-east-1, the region this offer is published for.
  • Availability stays available: the us-east-1 price list still lists an on-demand price, and the page says "you can continue to use a Legacy model during this period" (the Legacy period before the EOL date), though not as a new customer. The EOL date is in the future, so no retirement is recorded; reviewAfter is the EOL date, a re-review date, not a retirement.
  • Reviewed by Claude (owner delegation 2026-09-27), #909.

Raw API evidence ↗ · provider API · Claims: bedrock-minimax-m2-1: pricing · bedrock-minimax-m2-1: availability · 122 more claims about 61 other catalog entries · Retrieved · Checked · Review due · Confidence: A

  • Raw SHA-256: f1ac6a175fb27b3ad80d2c5d8970be8c2e5f957906349e52d15ec11d43e97896; collector aws-bedrock-price-list-json-1.
  • AWS Price List Bulk API offer file for Amazon Bedrock (us-east-1): offer version 20261111013037, publicationDate 2026-11-11T01:30:37Z.
  • Reviewed on-demand token prices only: each curated model's exact usagetypes in us-east-1, with service_tier "standard" for standard-tier models and with no service_tier attribute for the reviewed untiered models, in US dollars as the offer file states them (hence confidence A). A hosted listing and pricing observation only: no checkpoint, developer, license, context-window, modality or release-date assertions, and the usagetype is not a documented Bedrock model ID.
  • No Anthropic Claude price is published from this offer file: prices are read only from curated usagetypes, and no curated entry is an Anthropic model. (When the collector was reviewed on 2026-10-29, the file listed only legacy Claude models, all without a service tier.)
  • Availability rests on AWS's own offer file listing an on-demand price for the model in us-east-1.

MiniMax M2.5

Context window: 196K tokens

All rates, conditions and sources

Conditions of every rate below:

  • Listed for model "MiniMax M2.5" (provider "Minimax AI") in us-east-1 with service_tier "standard", as one USD price dimension from 0 to Inf.
  • AWS lists this price in USD per 1K tokens (unit "1K tokens"); it is multiplied by 1,000 exactly, by decimal arithmetic, to give USD per 1M tokens.
  • Only the standard on-demand tier in us-east-1 is published: flex, priority, batch and global cross-region tiers, other AWS Regions and provisioned throughput are excluded. The offer file lists no standard-tier cache price for this model; missing cache prices are unknown, not zero.

Input: $0.30 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-minimax.minimax-m2.5-mantle-input-tokens-standard.
  • Input tokens.

Output: $1.20 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-minimax.minimax-m2.5-mantle-output-tokens-standard.
  • Output tokens.

Price observed:

API availability: listed as available when checked (not a live service check)

Collection record

Primary source ↗ · provider documentation · Claims: bedrock-minimax-m2-5: specifications · Retrieved · Checked · Reviewed · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card ("Models at a glance") for MiniMax M2.5, Model ID minimax.minimax-m2.5 (the offer's price list usagetype stem), raw SHA-256 f4b8dde040f9c6d8b458524f1525c2daef6fc035e618d18ea2c4576f65e4ee84, read once on 2026-10-31 under the #837 AWS facts-plus-link basis: "Context window: 196K tokens".
  • Shorthand recorded as stated (#504); not converted to an exact token count. The value is AWS's statement for its Bedrock listing; another seller of the same model may state a different figure.
  • Specifications only: no pricing, availability, capability, lifecycle or release-date assertion.

Raw API evidence ↗ · provider API · Claims: bedrock-minimax-m2-5: pricing · bedrock-minimax-m2-5: availability · 122 more claims about 61 other catalog entries · Retrieved · Checked · Review due · Confidence: A

  • Raw SHA-256: f1ac6a175fb27b3ad80d2c5d8970be8c2e5f957906349e52d15ec11d43e97896; collector aws-bedrock-price-list-json-1.
  • AWS Price List Bulk API offer file for Amazon Bedrock (us-east-1): offer version 20261111013037, publicationDate 2026-11-11T01:30:37Z.
  • Reviewed on-demand token prices only: each curated model's exact usagetypes in us-east-1, with service_tier "standard" for standard-tier models and with no service_tier attribute for the reviewed untiered models, in US dollars as the offer file states them (hence confidence A). A hosted listing and pricing observation only: no checkpoint, developer, license, context-window, modality or release-date assertions, and the usagetype is not a documented Bedrock model ID.
  • No Anthropic Claude price is published from this offer file: prices are read only from curated usagetypes, and no curated entry is an Anthropic model. (When the collector was reviewed on 2026-10-29, the file listed only legacy Claude models, all without a service tier.)
  • Availability rests on AWS's own offer file listing an on-demand price for the model in us-east-1.

Ministral 14B 3.0

Context window: 128K tokens

All rates, conditions and sources

Conditions of every rate below:

  • Listed for model "Ministral 14B 3.0" (provider "Mistral") in us-east-1 with service_tier "standard", as one USD price dimension from 0 to Inf.
  • AWS lists this price in USD per 1K tokens (unit "1K tokens"); it is multiplied by 1,000 exactly, by decimal arithmetic, to give USD per 1M tokens.
  • Only the standard on-demand tier in us-east-1 is published: flex, priority, batch and global cross-region tiers, other AWS Regions and provisioned throughput are excluded. The offer file lists no standard-tier cache price for this model; missing cache prices are unknown, not zero.

Input: $0.20 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-mistral.ministral-3-14b-instruct-mantle-input-tokens-standard.
  • Input tokens.

Output: $0.20 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-mistral.ministral-3-14b-instruct-mantle-output-tokens-standard.
  • Output tokens.

Price observed:

API availability: listed as available when checked (not a live service check)

Collection record

Primary source ↗ · provider documentation · Claims: bedrock-ministral-3-14b-instruct: specifications · Retrieved · Checked · Reviewed · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card ("Models at a glance") for Ministral 14B 3.0, Model ID mistral.ministral-3-14b-instruct (the offer's price list usagetype stem), raw SHA-256 e6cd848b1c86de45b58c5d94e75b5ec4bf18db9545807a997a71dc63348889f5, read once on 2026-10-31 under the #837 AWS facts-plus-link basis: "Context window: 128K tokens".
  • Shorthand recorded as stated (#504); not converted to an exact token count. The value is AWS's statement for its Bedrock listing; another seller of the same model may state a different figure.
  • Specifications only: no pricing, availability, capability, lifecycle or release-date assertion.

Raw API evidence ↗ · provider API · Claims: bedrock-ministral-3-14b-instruct: pricing · bedrock-ministral-3-14b-instruct: availability · 122 more claims about 61 other catalog entries · Retrieved · Checked · Review due · Confidence: A

  • Raw SHA-256: f1ac6a175fb27b3ad80d2c5d8970be8c2e5f957906349e52d15ec11d43e97896; collector aws-bedrock-price-list-json-1.
  • AWS Price List Bulk API offer file for Amazon Bedrock (us-east-1): offer version 20261111013037, publicationDate 2026-11-11T01:30:37Z.
  • Reviewed on-demand token prices only: each curated model's exact usagetypes in us-east-1, with service_tier "standard" for standard-tier models and with no service_tier attribute for the reviewed untiered models, in US dollars as the offer file states them (hence confidence A). A hosted listing and pricing observation only: no checkpoint, developer, license, context-window, modality or release-date assertions, and the usagetype is not a documented Bedrock model ID.
  • No Anthropic Claude price is published from this offer file: prices are read only from curated usagetypes, and no curated entry is an Anthropic model. (When the collector was reviewed on 2026-10-29, the file listed only legacy Claude models, all without a service tier.)
  • Availability rests on AWS's own offer file listing an on-demand price for the model in us-east-1.

Ministral 3B 3.0

Context window: 128K tokens

All rates, conditions and sources

Conditions of every rate below:

  • Listed for model "Ministral 3B 3.0" (provider "Mistral") in us-east-1 with service_tier "standard", as one USD price dimension from 0 to Inf.
  • AWS lists this price in USD per 1K tokens (unit "1K tokens"); it is multiplied by 1,000 exactly, by decimal arithmetic, to give USD per 1M tokens.
  • Only the standard on-demand tier in us-east-1 is published: flex, priority, batch and global cross-region tiers, other AWS Regions and provisioned throughput are excluded. The offer file lists no standard-tier cache price for this model; missing cache prices are unknown, not zero.

Input: $0.10 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-mistral.ministral-3-3b-instruct-mantle-input-tokens-standard.
  • Input tokens.

Output: $0.10 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-mistral.ministral-3-3b-instruct-mantle-output-tokens-standard.
  • Output tokens.

Price observed:

API availability: listed as available when checked (not a live service check)

Collection record

Primary source ↗ · provider documentation · Claims: bedrock-ministral-3-3b-instruct: specifications · Retrieved · Checked · Reviewed · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card ("Models at a glance") for Ministral 3B, Model ID mistral.ministral-3-3b-instruct (the offer's price list usagetype stem), raw SHA-256 b5569a71e07f4e18134802d1a7459d8b64e7b5a4bf2a468423286697a0eb0cb8, read once on 2026-10-31 under the #837 AWS facts-plus-link basis: "Context window: 128K tokens".
  • Shorthand recorded as stated (#504); not converted to an exact token count. The value is AWS's statement for its Bedrock listing; another seller of the same model may state a different figure.
  • Specifications only: no pricing, availability, capability, lifecycle or release-date assertion.

Raw API evidence ↗ · provider API · Claims: bedrock-ministral-3-3b-instruct: pricing · bedrock-ministral-3-3b-instruct: availability · 122 more claims about 61 other catalog entries · Retrieved · Checked · Review due · Confidence: A

  • Raw SHA-256: f1ac6a175fb27b3ad80d2c5d8970be8c2e5f957906349e52d15ec11d43e97896; collector aws-bedrock-price-list-json-1.
  • AWS Price List Bulk API offer file for Amazon Bedrock (us-east-1): offer version 20261111013037, publicationDate 2026-11-11T01:30:37Z.
  • Reviewed on-demand token prices only: each curated model's exact usagetypes in us-east-1, with service_tier "standard" for standard-tier models and with no service_tier attribute for the reviewed untiered models, in US dollars as the offer file states them (hence confidence A). A hosted listing and pricing observation only: no checkpoint, developer, license, context-window, modality or release-date assertions, and the usagetype is not a documented Bedrock model ID.
  • No Anthropic Claude price is published from this offer file: prices are read only from curated usagetypes, and no curated entry is an Anthropic model. (When the collector was reviewed on 2026-10-29, the file listed only legacy Claude models, all without a service tier.)
  • Availability rests on AWS's own offer file listing an on-demand price for the model in us-east-1.

Ministral 8B 3.0

Context window: 128K tokens

All rates, conditions and sources

Conditions of every rate below:

  • Listed for model "Ministral 8B 3.0" (provider "Mistral") in us-east-1 with service_tier "standard", as one USD price dimension from 0 to Inf.
  • AWS lists this price in USD per 1K tokens (unit "1K tokens"); it is multiplied by 1,000 exactly, by decimal arithmetic, to give USD per 1M tokens.
  • Only the standard on-demand tier in us-east-1 is published: flex, priority, batch and global cross-region tiers, other AWS Regions and provisioned throughput are excluded. The offer file lists no standard-tier cache price for this model; missing cache prices are unknown, not zero.

Input: $0.15 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-mistral.ministral-3-8b-instruct-mantle-input-tokens-standard.
  • Input tokens.

Output: $0.15 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-mistral.ministral-3-8b-instruct-mantle-output-tokens-standard.
  • Output tokens.

Price observed:

API availability: listed as available when checked (not a live service check)

Collection record

Primary source ↗ · provider documentation · Claims: bedrock-ministral-3-8b-instruct: specifications · Retrieved · Checked · Reviewed · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card ("Models at a glance") for Ministral 3 8B, Model ID mistral.ministral-3-8b-instruct (the offer's price list usagetype stem), raw SHA-256 2fb549954fab44c482a35f15f055955da03bb7f49fd56ab5fdf3b28c13b8cbd0, read once on 2026-10-31 under the #837 AWS facts-plus-link basis: "Context window: 128K tokens".
  • Shorthand recorded as stated (#504); not converted to an exact token count. The value is AWS's statement for its Bedrock listing; another seller of the same model may state a different figure.
  • Specifications only: no pricing, availability, capability, lifecycle or release-date assertion.

Raw API evidence ↗ · provider API · Claims: bedrock-ministral-3-8b-instruct: pricing · bedrock-ministral-3-8b-instruct: availability · 122 more claims about 61 other catalog entries · Retrieved · Checked · Review due · Confidence: A

  • Raw SHA-256: f1ac6a175fb27b3ad80d2c5d8970be8c2e5f957906349e52d15ec11d43e97896; collector aws-bedrock-price-list-json-1.
  • AWS Price List Bulk API offer file for Amazon Bedrock (us-east-1): offer version 20261111013037, publicationDate 2026-11-11T01:30:37Z.
  • Reviewed on-demand token prices only: each curated model's exact usagetypes in us-east-1, with service_tier "standard" for standard-tier models and with no service_tier attribute for the reviewed untiered models, in US dollars as the offer file states them (hence confidence A). A hosted listing and pricing observation only: no checkpoint, developer, license, context-window, modality or release-date assertions, and the usagetype is not a documented Bedrock model ID.
  • No Anthropic Claude price is published from this offer file: prices are read only from curated usagetypes, and no curated entry is an Anthropic model. (When the collector was reviewed on 2026-10-29, the file listed only legacy Claude models, all without a service tier.)
  • Availability rests on AWS's own offer file listing an on-demand price for the model in us-east-1.

Mistral 7B

Deprecated by provider · still listed · price checked Nov 12, 2026 (deprecation notice ↗)

All rates, conditions and sources

Conditions of every rate below:

  • Listed for model "Mistral 7B" (provider "Mistral") in us-east-1 with no service_tier attribute (feature "On-demand Inference"), as one USD price dimension from 0 to Inf.
  • AWS lists this price in USD per 1K tokens (unit "1K tokens"); it is multiplied by 1,000 exactly, by decimal arithmetic, to give USD per 1M tokens.
  • AWS's on-demand price for this model with no service tier stated: the row carries no service_tier attribute, so it is not labelled as the standard or any other tier. Only this plain on-demand usagetype in us-east-1 is published: any flex, priority, batch, latency-optimized, custom-model or cross-region (global) variant, other AWS Regions, provisioned throughput and model customization are excluded. The offer file lists no cache price for this model without a service tier; missing cache prices are unknown, not zero.

Input: $0.15 per 1M tokens · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-Mistral7B-input-tokens.
  • Input tokens.

Output: $0.20 per 1M tokens · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-Mistral7B-output-tokens.
  • Output tokens.

Price observed:

API availability: listed as available when checked (not a live service check)

Collection record

Primary source ↗ · provider documentation · Claims: bedrock-mistral-7b: availability · Retrieved · Checked · Reviewed · Review due · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card for Mistral 7B Instruct, Model ID mistral.mistral-7b-instruct-v0:2, raw SHA-256 0d5f00e268659191bcf6f7eeb4cfb2f75d6de4d394cbf540a45e2e191f5c3670, read once at 2026-11-02T11:40:22Z under the #837 AWS facts-plus-link basis: "Model lifecycle: Legacy"; "Model EOL date: March 30, 2027"; "Model lifecycle policy: Model lifecycle (For Models Launched Prior to Sept 7 2026)".
  • Amazon Bedrock "Model lifecycle (Legacy)" page (https://docs.aws.amazon.com/bedrock/latest/userguide/model-lifecycle-legacy.html), read once at 2026-11-02T11:37:59Z, raw SHA-256 c33d9d25d7da7f47c098aa864830361de770224f38a790070bada021dac56850: its Legacy table lists Mistral 7B Instruct (mistral.mistral-7b-instruct-v0:2) with Legacy date September 30, 2026, EOL date March 30, 2027 and public extended access start date "To be announced". The page states: "New customers can't use Legacy models and existing customers may lose access to Legacy models after 15 days of inactivity."
  • That table row's Regions (ap-south-1, ap-southeast-2, ca-central-1, eu-west-1, eu-west-2, eu-west-3, sa-east-1, us-east-1, us-west-2) include us-east-1, the region this offer is published for.
  • Availability stays available: the us-east-1 price list still lists an on-demand price (2026-11-06 archive), and the page says "you can continue to use a Legacy model during this period" (the Legacy period before the EOL date), though not as a new customer. The EOL date is in the future, so no retirement is recorded; reviewAfter is the EOL date, a re-review date, not a retirement.
  • Reviewed by Claude (owner delegation 2026-09-27), #863.

Raw API evidence ↗ · provider API · Claims: bedrock-mistral-7b: pricing · bedrock-mistral-7b: availability · 122 more claims about 61 other catalog entries · Retrieved · Checked · Review due · Confidence: A

  • Raw SHA-256: f1ac6a175fb27b3ad80d2c5d8970be8c2e5f957906349e52d15ec11d43e97896; collector aws-bedrock-price-list-json-1.
  • AWS Price List Bulk API offer file for Amazon Bedrock (us-east-1): offer version 20261111013037, publicationDate 2026-11-11T01:30:37Z.
  • Reviewed on-demand token prices only: each curated model's exact usagetypes in us-east-1, with service_tier "standard" for standard-tier models and with no service_tier attribute for the reviewed untiered models, in US dollars as the offer file states them (hence confidence A). A hosted listing and pricing observation only: no checkpoint, developer, license, context-window, modality or release-date assertions, and the usagetype is not a documented Bedrock model ID.
  • No Anthropic Claude price is published from this offer file: prices are read only from curated usagetypes, and no curated entry is an Anthropic model. (When the collector was reviewed on 2026-10-29, the file listed only legacy Claude models, all without a service tier.)
  • Availability rests on AWS's own offer file listing an on-demand price for the model in us-east-1.

Mistral Large

Context window: 32K tokens

All rates, conditions and sources

Conditions of every rate below:

  • Listed for model "Mistral Large" (provider "Mistral") in us-east-1 with no service_tier attribute (feature "On-demand Inference"), as one USD price dimension from 0 to Inf.
  • AWS lists this price in USD per 1K tokens (unit "1K tokens"); it is multiplied by 1,000 exactly, by decimal arithmetic, to give USD per 1M tokens.
  • AWS's on-demand price for this model with no service tier stated: the row carries no service_tier attribute, so it is not labelled as the standard or any other tier. Only this plain on-demand usagetype in us-east-1 is published: any flex, priority, batch, latency-optimized, custom-model or cross-region (global) variant, other AWS Regions, provisioned throughput and model customization are excluded. The offer file lists no cache price for this model without a service tier; missing cache prices are unknown, not zero.

Input: $4.00 per 1M tokens · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-MistralLarge-input-tokens.
  • Input tokens.

Output: $12.00 per 1M tokens · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-MistralLarge-output-tokens.
  • Output tokens.

Price observed:

API availability: listed as available when checked (not a live service check)

Collection record

Primary source ↗ · provider documentation · Claims: bedrock-mistral-large: specifications · Retrieved · Checked · Reviewed · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card ("Models at a glance") for Mistral Large, Model ID mistral.mistral-large-2402-v1:0. The offer's price list usagetype (USE1-MistralLarge-input-tokens) is a price list name, not a model ID, so identity rests on the price list model attribute "Mistral Large" matching this card. The index's only other Mistral Large card is Mistral Large 3, which has its own standard-tier offer (bedrock-mistral-large-3-675b-instruct). Raw SHA-256 faa21bd2dcb3f883bfc967c83e7135c7608103676a7fa30f73698b23e32d0605, read once on 2026-11-07 under the #837 AWS facts-plus-link basis: "Context window: 32K tokens".
  • Shorthand recorded as stated (#504); not converted to an exact token count. The value is AWS's statement for its Bedrock listing; another seller of the same model may state a different figure.
  • Specifications only: no pricing, availability, capability, lifecycle or release-date assertion.

Raw API evidence ↗ · provider API · Claims: bedrock-mistral-large: pricing · bedrock-mistral-large: availability · 122 more claims about 61 other catalog entries · Retrieved · Checked · Review due · Confidence: A

  • Raw SHA-256: f1ac6a175fb27b3ad80d2c5d8970be8c2e5f957906349e52d15ec11d43e97896; collector aws-bedrock-price-list-json-1.
  • AWS Price List Bulk API offer file for Amazon Bedrock (us-east-1): offer version 20261111013037, publicationDate 2026-11-11T01:30:37Z.
  • Reviewed on-demand token prices only: each curated model's exact usagetypes in us-east-1, with service_tier "standard" for standard-tier models and with no service_tier attribute for the reviewed untiered models, in US dollars as the offer file states them (hence confidence A). A hosted listing and pricing observation only: no checkpoint, developer, license, context-window, modality or release-date assertions, and the usagetype is not a documented Bedrock model ID.
  • No Anthropic Claude price is published from this offer file: prices are read only from curated usagetypes, and no curated entry is an Anthropic model. (When the collector was reviewed on 2026-10-29, the file listed only legacy Claude models, all without a service tier.)
  • Availability rests on AWS's own offer file listing an on-demand price for the model in us-east-1.

Mistral Large 3

Context window: 256K tokens

All rates, conditions and sources

Conditions of every rate below:

  • Listed for model "Mistral Large 3" (provider "Mistral") in us-east-1 with service_tier "standard", as one USD price dimension from 0 to Inf.
  • AWS lists this price in USD per 1K tokens (unit "1K tokens"); it is multiplied by 1,000 exactly, by decimal arithmetic, to give USD per 1M tokens.
  • Only the standard on-demand tier in us-east-1 is published: flex, priority, batch and global cross-region tiers, other AWS Regions and provisioned throughput are excluded. The offer file lists no standard-tier cache price for this model; missing cache prices are unknown, not zero.

Input: $0.50 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-mistral.mistral-large-3-675b-instruct-mantle-input-tokens-standard.
  • Input tokens.

Output: $1.50 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-mistral.mistral-large-3-675b-instruct-mantle-output-tokens-standard.
  • Output tokens.

Price observed:

API availability: listed as available when checked (not a live service check)

Collection record

Primary source ↗ · provider documentation · Claims: bedrock-mistral-large-3-675b-instruct: specifications · Retrieved · Checked · Reviewed · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card ("Models at a glance") for Mistral Large 3, Model ID mistral.mistral-large-3-675b-instruct (the offer's price list usagetype stem), raw SHA-256 459c968ffedea61504aa8bafaecc99696db5b0a25ff6cdd1a6f3a5bb1bfbfa76, read once on 2026-10-31 under the #837 AWS facts-plus-link basis: "Context window: 256K tokens".
  • Shorthand recorded as stated (#504); not converted to an exact token count. The value is AWS's statement for its Bedrock listing; another seller of the same model may state a different figure.
  • Specifications only: no pricing, availability, capability, lifecycle or release-date assertion.

Raw API evidence ↗ · provider API · Claims: bedrock-mistral-large-3-675b-instruct: pricing · bedrock-mistral-large-3-675b-instruct: availability · 122 more claims about 61 other catalog entries · Retrieved · Checked · Review due · Confidence: A

  • Raw SHA-256: f1ac6a175fb27b3ad80d2c5d8970be8c2e5f957906349e52d15ec11d43e97896; collector aws-bedrock-price-list-json-1.
  • AWS Price List Bulk API offer file for Amazon Bedrock (us-east-1): offer version 20261111013037, publicationDate 2026-11-11T01:30:37Z.
  • Reviewed on-demand token prices only: each curated model's exact usagetypes in us-east-1, with service_tier "standard" for standard-tier models and with no service_tier attribute for the reviewed untiered models, in US dollars as the offer file states them (hence confidence A). A hosted listing and pricing observation only: no checkpoint, developer, license, context-window, modality or release-date assertions, and the usagetype is not a documented Bedrock model ID.
  • No Anthropic Claude price is published from this offer file: prices are read only from curated usagetypes, and no curated entry is an Anthropic model. (When the collector was reviewed on 2026-10-29, the file listed only legacy Claude models, all without a service tier.)
  • Availability rests on AWS's own offer file listing an on-demand price for the model in us-east-1.

Mistral Small

Context window: 32K tokens

All rates, conditions and sources

Conditions of every rate below:

  • Listed for model "Mistral Small" (provider "Mistral") in us-east-1 with no service_tier attribute (feature "On-demand Inference"), as one USD price dimension from 0 to Inf.
  • AWS lists this price in USD per 1K tokens (unit "1K tokens"); it is multiplied by 1,000 exactly, by decimal arithmetic, to give USD per 1M tokens.
  • AWS's on-demand price for this model with no service tier stated: the row carries no service_tier attribute, so it is not labelled as the standard or any other tier. Only this plain on-demand usagetype in us-east-1 is published: any flex, priority, batch, latency-optimized, custom-model or cross-region (global) variant, other AWS Regions, provisioned throughput and model customization are excluded. The offer file lists no cache price for this model without a service tier; missing cache prices are unknown, not zero.

Input: $1.00 per 1M tokens · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-MistralSmall-input-tokens.
  • Input tokens.

Output: $3.00 per 1M tokens · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-MistralSmall-output-tokens.
  • Output tokens.

Price observed:

API availability: listed as available when checked (not a live service check)

Collection record

Primary source ↗ · provider documentation · Claims: bedrock-mistral-small: specifications · Retrieved · Checked · Reviewed · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card ("Models at a glance") for Mistral Small, Model ID mistral.mistral-small-2402-v1:0. The offer's price list usagetype (USE1-MistralSmall-input-tokens) is a price list name, not a model ID, so identity rests on the price list model attribute "Mistral Small" matching this card. Raw SHA-256 cd6919d31d743130a3e9c9e8ee271d70f8342d5be55171061cf6fcbd160e98a3, read once on 2026-11-07 under the #837 AWS facts-plus-link basis: "Context window: 32K tokens".
  • Shorthand recorded as stated (#504); not converted to an exact token count. The value is AWS's statement for its Bedrock listing; another seller of the same model may state a different figure.
  • Specifications only: no pricing, availability, capability, lifecycle or release-date assertion.

Raw API evidence ↗ · provider API · Claims: bedrock-mistral-small: pricing · bedrock-mistral-small: availability · 122 more claims about 61 other catalog entries · Retrieved · Checked · Review due · Confidence: A

  • Raw SHA-256: f1ac6a175fb27b3ad80d2c5d8970be8c2e5f957906349e52d15ec11d43e97896; collector aws-bedrock-price-list-json-1.
  • AWS Price List Bulk API offer file for Amazon Bedrock (us-east-1): offer version 20261111013037, publicationDate 2026-11-11T01:30:37Z.
  • Reviewed on-demand token prices only: each curated model's exact usagetypes in us-east-1, with service_tier "standard" for standard-tier models and with no service_tier attribute for the reviewed untiered models, in US dollars as the offer file states them (hence confidence A). A hosted listing and pricing observation only: no checkpoint, developer, license, context-window, modality or release-date assertions, and the usagetype is not a documented Bedrock model ID.
  • No Anthropic Claude price is published from this offer file: prices are read only from curated usagetypes, and no curated entry is an Anthropic model. (When the collector was reviewed on 2026-10-29, the file listed only legacy Claude models, all without a service tier.)
  • Availability rests on AWS's own offer file listing an on-demand price for the model in us-east-1.

Mixtral 8x7B

Deprecated by provider · still listed · price checked Nov 12, 2026 (deprecation notice ↗)

All rates, conditions and sources

Conditions of every rate below:

  • Listed for model "Mixtral 8x7B" (provider "Mistral") in us-east-1 with no service_tier attribute (feature "On-demand Inference"), as one USD price dimension from 0 to Inf.
  • AWS lists this price in USD per 1K tokens (unit "1K tokens"); it is multiplied by 1,000 exactly, by decimal arithmetic, to give USD per 1M tokens.
  • AWS's on-demand price for this model with no service tier stated: the row carries no service_tier attribute, so it is not labelled as the standard or any other tier. Only this plain on-demand usagetype in us-east-1 is published: any flex, priority, batch, latency-optimized, custom-model or cross-region (global) variant, other AWS Regions, provisioned throughput and model customization are excluded. The offer file lists no cache price for this model without a service tier; missing cache prices are unknown, not zero.

Input: $0.45 per 1M tokens · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-Mixtral8x7B-input-tokens.
  • Input tokens.

Output: $0.70 per 1M tokens · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-Mixtral8x7B-output-tokens.
  • Output tokens.

Price observed:

API availability: listed as available when checked (not a live service check)

Collection record

Primary source ↗ · provider documentation · Claims: bedrock-mixtral-8x7b: availability · Retrieved · Checked · Reviewed · Review due · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card for Mixtral 8x7B Instruct, Model ID mistral.mixtral-8x7b-instruct-v0:1, raw SHA-256 bde8f44bcffd85d6352459f110cba46467595b4bd570a9ef7a5ac3cbd5a40bbb, read once at 2026-11-02T11:40:29Z under the #837 AWS facts-plus-link basis: "Model lifecycle: Legacy"; "Model EOL date: March 30, 2027"; "Model lifecycle policy: Model lifecycle (For Models Launched Prior to Sept 7 2026)".
  • Amazon Bedrock "Model lifecycle (Legacy)" page (https://docs.aws.amazon.com/bedrock/latest/userguide/model-lifecycle-legacy.html), read once at 2026-11-02T11:37:59Z, raw SHA-256 c33d9d25d7da7f47c098aa864830361de770224f38a790070bada021dac56850: its Legacy table lists Mixtral 8x7B Instruct (mistral.mixtral-8x7b-instruct-v0:1) with Legacy date September 30, 2026, EOL date March 30, 2027 and public extended access start date "To be announced". The page states: "New customers can't use Legacy models and existing customers may lose access to Legacy models after 15 days of inactivity."
  • That table row's Regions (ap-south-1, ap-southeast-2, ca-central-1, eu-west-1, eu-west-2, eu-west-3, sa-east-1, us-east-1, us-west-2) include us-east-1, the region this offer is published for.
  • Availability stays available: the us-east-1 price list still lists an on-demand price (2026-11-06 archive), and the page says "you can continue to use a Legacy model during this period" (the Legacy period before the EOL date), though not as a new customer. The EOL date is in the future, so no retirement is recorded; reviewAfter is the EOL date, a re-review date, not a retirement.
  • Reviewed by Claude (owner delegation 2026-09-27), #863.

Raw API evidence ↗ · provider API · Claims: bedrock-mixtral-8x7b: pricing · bedrock-mixtral-8x7b: availability · 122 more claims about 61 other catalog entries · Retrieved · Checked · Review due · Confidence: A

  • Raw SHA-256: f1ac6a175fb27b3ad80d2c5d8970be8c2e5f957906349e52d15ec11d43e97896; collector aws-bedrock-price-list-json-1.
  • AWS Price List Bulk API offer file for Amazon Bedrock (us-east-1): offer version 20261111013037, publicationDate 2026-11-11T01:30:37Z.
  • Reviewed on-demand token prices only: each curated model's exact usagetypes in us-east-1, with service_tier "standard" for standard-tier models and with no service_tier attribute for the reviewed untiered models, in US dollars as the offer file states them (hence confidence A). A hosted listing and pricing observation only: no checkpoint, developer, license, context-window, modality or release-date assertions, and the usagetype is not a documented Bedrock model ID.
  • No Anthropic Claude price is published from this offer file: prices are read only from curated usagetypes, and no curated entry is an Anthropic model. (When the collector was reviewed on 2026-10-29, the file listed only legacy Claude models, all without a service tier.)
  • Availability rests on AWS's own offer file listing an on-demand price for the model in us-east-1.

Nemotron Nano 3 30B

Context window: 256K tokens

All rates, conditions and sources

Conditions of every rate below:

  • Listed for model "Nemotron Nano 3 30B" (provider "Nvidia") in us-east-1 with service_tier "standard", as one USD price dimension from 0 to Inf.
  • AWS lists this price in USD per 1K tokens (unit "1K tokens"); it is multiplied by 1,000 exactly, by decimal arithmetic, to give USD per 1M tokens.
  • Only the standard on-demand tier in us-east-1 is published: flex, priority, batch and global cross-region tiers, other AWS Regions and provisioned throughput are excluded. The offer file lists no standard-tier cache price for this model; missing cache prices are unknown, not zero.

Input: $0.06 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-nvidia.nemotron-nano-3-30b-mantle-input-tokens-standard.
  • Input tokens.

Output: $0.24 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-nvidia.nemotron-nano-3-30b-mantle-output-tokens-standard.
  • Output tokens.

Price observed:

API availability: listed as available when checked (not a live service check)

Collection record

Primary source ↗ · provider documentation · Claims: bedrock-nemotron-nano-3-30b: specifications · Retrieved · Checked · Reviewed · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card ("Models at a glance") for Nemotron Nano 3 30B, Model ID nvidia.nemotron-nano-3-30b (the offer's price list usagetype stem), raw SHA-256 0a04c2cf62d57710cfe442bb4eb104775964fafeb5ad89f9fd9d14b640dc5596, read once on 2026-10-31 under the #837 AWS facts-plus-link basis: "Context window: 256K tokens".
  • Shorthand recorded as stated (#504); not converted to an exact token count. The value is AWS's statement for its Bedrock listing; another seller of the same model may state a different figure.
  • Specifications only: no pricing, availability, capability, lifecycle or release-date assertion.

Raw API evidence ↗ · provider API · Claims: bedrock-nemotron-nano-3-30b: pricing · bedrock-nemotron-nano-3-30b: availability · 122 more claims about 61 other catalog entries · Retrieved · Checked · Review due · Confidence: A

  • Raw SHA-256: f1ac6a175fb27b3ad80d2c5d8970be8c2e5f957906349e52d15ec11d43e97896; collector aws-bedrock-price-list-json-1.
  • AWS Price List Bulk API offer file for Amazon Bedrock (us-east-1): offer version 20261111013037, publicationDate 2026-11-11T01:30:37Z.
  • Reviewed on-demand token prices only: each curated model's exact usagetypes in us-east-1, with service_tier "standard" for standard-tier models and with no service_tier attribute for the reviewed untiered models, in US dollars as the offer file states them (hence confidence A). A hosted listing and pricing observation only: no checkpoint, developer, license, context-window, modality or release-date assertions, and the usagetype is not a documented Bedrock model ID.
  • No Anthropic Claude price is published from this offer file: prices are read only from curated usagetypes, and no curated entry is an Anthropic model. (When the collector was reviewed on 2026-10-29, the file listed only legacy Claude models, all without a service tier.)
  • Availability rests on AWS's own offer file listing an on-demand price for the model in us-east-1.

Nova 2.0 Lite

Context window: 1M tokens

All rates, conditions and sources

Conditions of every rate below:

  • Listed for model "Nova 2.0 Lite" in us-east-1 with no service_tier attribute (feature "On-demand Inference"), as one USD price dimension from 0 to Inf.
  • AWS lists this price in USD per 1K tokens (unit "1K tokens"); it is multiplied by 1,000 exactly, by decimal arithmetic, to give USD per 1M tokens.
  • AWS's on-demand price for this model with no service tier stated: the row carries no service_tier attribute, so it is not labelled as the standard or any other tier. Only this plain on-demand usagetype in us-east-1 is published: any flex, priority, batch, latency-optimized, custom-model or cross-region (global) variant, other AWS Regions, provisioned throughput and model customization are excluded. The offer file lists a cache read price for this model as a separate usagetype; cache write usagetypes are not published.

Input: $0.33 per 1M tokens · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-Nova2.0Lite-input-tokens.
  • Input tokens.

Output: $2.75 per 1M tokens · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-Nova2.0Lite-output-tokens.
  • Output tokens.

Cache read: $0.0825 per 1M tokens · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-Nova2.0Lite-cache-read-input-token-count.
  • Cache read tokens (a "cache-read-input-token-count" usagetype). The offer file states no cache eligibility or cache lifetime for reads.

Price observed:

API availability: listed as available when checked (not a live service check)

Collection record

Primary source ↗ · provider documentation · Claims: bedrock-nova-2-0-lite: specifications · Retrieved · Checked · Reviewed · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card ("Models at a glance") for Nova 2 Lite, Model ID amazon.nova-2-lite-v1:0. The offer's price list usagetype (USE1-Nova2.0Lite-input-tokens) is a price list name, not a model ID, so identity rests on the price list model attribute "Nova 2.0 Lite" matching this card. The attribute writes "2.0" where the card title writes "2"; this is the index's only Nova 2 Lite card, and its Model ID is nova-2-lite. Raw SHA-256 49df493d7d5431cbb3d99aa7ce645f37e23446d84fe6a1ebacf2383ec9fbb9df, read once on 2026-11-07 under the #837 AWS facts-plus-link basis: "Context window: 1M tokens".
  • Shorthand recorded as stated (#504); not converted to an exact token count. The value is AWS's statement for its Bedrock listing; another seller of the same model may state a different figure.
  • Specifications only: no pricing, availability, capability, lifecycle or release-date assertion.

Raw API evidence ↗ · provider API · Claims: bedrock-nova-2-0-lite: pricing · bedrock-nova-2-0-lite: availability · 122 more claims about 61 other catalog entries · Retrieved · Checked · Review due · Confidence: A

  • Raw SHA-256: f1ac6a175fb27b3ad80d2c5d8970be8c2e5f957906349e52d15ec11d43e97896; collector aws-bedrock-price-list-json-1.
  • AWS Price List Bulk API offer file for Amazon Bedrock (us-east-1): offer version 20261111013037, publicationDate 2026-11-11T01:30:37Z.
  • Reviewed on-demand token prices only: each curated model's exact usagetypes in us-east-1, with service_tier "standard" for standard-tier models and with no service_tier attribute for the reviewed untiered models, in US dollars as the offer file states them (hence confidence A). A hosted listing and pricing observation only: no checkpoint, developer, license, context-window, modality or release-date assertions, and the usagetype is not a documented Bedrock model ID.
  • No Anthropic Claude price is published from this offer file: prices are read only from curated usagetypes, and no curated entry is an Anthropic model. (When the collector was reviewed on 2026-10-29, the file listed only legacy Claude models, all without a service tier.)
  • Availability rests on AWS's own offer file listing an on-demand price for the model in us-east-1.

Nova Lite

Context window: 300K tokens

All rates, conditions and sources

Conditions of every rate below:

  • Listed for model "Nova Lite" in us-east-1 with no service_tier attribute (feature "On-demand Inference"), as one USD price dimension from 0 to Inf.
  • AWS lists this price in USD per 1K tokens (unit "1K tokens"); it is multiplied by 1,000 exactly, by decimal arithmetic, to give USD per 1M tokens.
  • AWS's on-demand price for this model with no service tier stated: the row carries no service_tier attribute, so it is not labelled as the standard or any other tier. Only this plain on-demand usagetype in us-east-1 is published: any flex, priority, batch, latency-optimized, custom-model or cross-region (global) variant, other AWS Regions, provisioned throughput and model customization are excluded. The offer file lists a cache read price for this model as a separate usagetype; cache write usagetypes are not published.

Input: $0.06 per 1M tokens · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-NovaLite-input-tokens.
  • Input tokens.

Output: $0.24 per 1M tokens · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-NovaLite-output-tokens.
  • Output tokens.

Cache read: $0.015 per 1M tokens · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-NovaLite-cache-read-input-token-count.
  • Cache read tokens (a "cache-read-input-token-count" usagetype). The offer file states no cache eligibility or cache lifetime for reads.

Price observed:

API availability: listed as available when checked (not a live service check)

Collection record

Primary source ↗ · provider documentation · Claims: bedrock-nova-lite: specifications · Retrieved · Checked · Reviewed · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card ("Models at a glance") for Nova Lite, Model ID amazon.nova-lite-v1:0. The offer's price list usagetype (USE1-NovaLite-input-tokens) is a price list name, not a model ID, so identity rests on the price list model attribute "Nova Lite" matching this card. Raw SHA-256 7a1a06e5c6f6c110327785d039277bbdf53d3204fc4796910213b58cdccbd6a0, read once on 2026-11-07 under the #837 AWS facts-plus-link basis: "Context window: 300K tokens".
  • Shorthand recorded as stated (#504); not converted to an exact token count. The value is AWS's statement for its Bedrock listing; another seller of the same model may state a different figure.
  • Specifications only: no pricing, availability, capability, lifecycle or release-date assertion.

Raw API evidence ↗ · provider API · Claims: bedrock-nova-lite: pricing · bedrock-nova-lite: availability · 122 more claims about 61 other catalog entries · Retrieved · Checked · Review due · Confidence: A

  • Raw SHA-256: f1ac6a175fb27b3ad80d2c5d8970be8c2e5f957906349e52d15ec11d43e97896; collector aws-bedrock-price-list-json-1.
  • AWS Price List Bulk API offer file for Amazon Bedrock (us-east-1): offer version 20261111013037, publicationDate 2026-11-11T01:30:37Z.
  • Reviewed on-demand token prices only: each curated model's exact usagetypes in us-east-1, with service_tier "standard" for standard-tier models and with no service_tier attribute for the reviewed untiered models, in US dollars as the offer file states them (hence confidence A). A hosted listing and pricing observation only: no checkpoint, developer, license, context-window, modality or release-date assertions, and the usagetype is not a documented Bedrock model ID.
  • No Anthropic Claude price is published from this offer file: prices are read only from curated usagetypes, and no curated entry is an Anthropic model. (When the collector was reviewed on 2026-10-29, the file listed only legacy Claude models, all without a service tier.)
  • Availability rests on AWS's own offer file listing an on-demand price for the model in us-east-1.

Nova Micro

Context window: 128K tokens

All rates, conditions and sources

Conditions of every rate below:

  • Listed for model "Nova Micro" in us-east-1 with no service_tier attribute (feature "On-demand Inference"), as one USD price dimension from 0 to Inf.
  • AWS lists this price in USD per 1K tokens (unit "1K tokens"); it is multiplied by 1,000 exactly, by decimal arithmetic, to give USD per 1M tokens.
  • AWS's on-demand price for this model with no service tier stated: the row carries no service_tier attribute, so it is not labelled as the standard or any other tier. Only this plain on-demand usagetype in us-east-1 is published: any flex, priority, batch, latency-optimized, custom-model or cross-region (global) variant, other AWS Regions, provisioned throughput and model customization are excluded. The offer file lists a cache read price for this model as a separate usagetype; cache write usagetypes are not published.

Input: $0.035 per 1M tokens · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-NovaMicro-input-tokens.
  • Input tokens.

Output: $0.14 per 1M tokens · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-NovaMicro-output-tokens.
  • Output tokens.

Cache read: $0.00875 per 1M tokens · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-NovaMicro-cache-read-input-token-count.
  • Cache read tokens (a "cache-read-input-token-count" usagetype). The offer file states no cache eligibility or cache lifetime for reads.

Price observed:

API availability: listed as available when checked (not a live service check)

Collection record

Primary source ↗ · provider documentation · Claims: bedrock-nova-micro: specifications · Retrieved · Checked · Reviewed · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card ("Models at a glance") for Nova Micro, Model ID amazon.nova-micro-v1:0. The offer's price list usagetype (USE1-NovaMicro-input-tokens) is a price list name, not a model ID, so identity rests on the price list model attribute "Nova Micro" matching this card. Raw SHA-256 7128161981cf9fbaa9827974dff27e22fdcc50dbb0da5ccc0e8a79d73d590bee, read once on 2026-11-07 under the #837 AWS facts-plus-link basis: "Context window: 128K tokens".
  • Shorthand recorded as stated (#504); not converted to an exact token count. The value is AWS's statement for its Bedrock listing; another seller of the same model may state a different figure.
  • Specifications only: no pricing, availability, capability, lifecycle or release-date assertion.

Raw API evidence ↗ · provider API · Claims: bedrock-nova-micro: pricing · bedrock-nova-micro: availability · 122 more claims about 61 other catalog entries · Retrieved · Checked · Review due · Confidence: A

  • Raw SHA-256: f1ac6a175fb27b3ad80d2c5d8970be8c2e5f957906349e52d15ec11d43e97896; collector aws-bedrock-price-list-json-1.
  • AWS Price List Bulk API offer file for Amazon Bedrock (us-east-1): offer version 20261111013037, publicationDate 2026-11-11T01:30:37Z.
  • Reviewed on-demand token prices only: each curated model's exact usagetypes in us-east-1, with service_tier "standard" for standard-tier models and with no service_tier attribute for the reviewed untiered models, in US dollars as the offer file states them (hence confidence A). A hosted listing and pricing observation only: no checkpoint, developer, license, context-window, modality or release-date assertions, and the usagetype is not a documented Bedrock model ID.
  • No Anthropic Claude price is published from this offer file: prices are read only from curated usagetypes, and no curated entry is an Anthropic model. (When the collector was reviewed on 2026-10-29, the file listed only legacy Claude models, all without a service tier.)
  • Availability rests on AWS's own offer file listing an on-demand price for the model in us-east-1.

Nova Premier

All rates, conditions and sources

Conditions of every rate below:

  • Listed for model "Nova Premier" in us-east-1 with no service_tier attribute (feature "On-demand Inference"), as one USD price dimension from 0 to Inf.
  • AWS lists this price in USD per 1K tokens (unit "1K tokens"); it is multiplied by 1,000 exactly, by decimal arithmetic, to give USD per 1M tokens.
  • AWS's on-demand price for this model with no service tier stated: the row carries no service_tier attribute, so it is not labelled as the standard or any other tier. Only this plain on-demand usagetype in us-east-1 is published: any flex, priority, batch, latency-optimized, custom-model or cross-region (global) variant, other AWS Regions, provisioned throughput and model customization are excluded. The offer file lists a cache read price for this model as a separate usagetype; cache write usagetypes are not published.

Input: $2.50 per 1M tokens · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-NovaPremier-input-tokens.
  • Input tokens.

Output: $12.50 per 1M tokens · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-NovaPremier-output-tokens.
  • Output tokens.

Cache read: $0.625 per 1M tokens · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-NovaPremier-cache-read-input-token-count.
  • Cache read tokens (a "cache-read-input-token-count" usagetype). The offer file states no cache eligibility or cache lifetime for reads.

Price observed:

API availability: listed as available when checked (not a live service check)

Collection record

Raw API evidence ↗ · provider API · Claims: bedrock-nova-premier: pricing · bedrock-nova-premier: availability · 122 more claims about 61 other catalog entries · Retrieved · Checked · Review due · Confidence: A

  • Raw SHA-256: f1ac6a175fb27b3ad80d2c5d8970be8c2e5f957906349e52d15ec11d43e97896; collector aws-bedrock-price-list-json-1.
  • AWS Price List Bulk API offer file for Amazon Bedrock (us-east-1): offer version 20261111013037, publicationDate 2026-11-11T01:30:37Z.
  • Reviewed on-demand token prices only: each curated model's exact usagetypes in us-east-1, with service_tier "standard" for standard-tier models and with no service_tier attribute for the reviewed untiered models, in US dollars as the offer file states them (hence confidence A). A hosted listing and pricing observation only: no checkpoint, developer, license, context-window, modality or release-date assertions, and the usagetype is not a documented Bedrock model ID.
  • No Anthropic Claude price is published from this offer file: prices are read only from curated usagetypes, and no curated entry is an Anthropic model. (When the collector was reviewed on 2026-10-29, the file listed only legacy Claude models, all without a service tier.)
  • Availability rests on AWS's own offer file listing an on-demand price for the model in us-east-1.

Nova Pro

Context window: 300K tokens

All rates, conditions and sources

Conditions of every rate below:

  • Listed for model "Nova Pro" in us-east-1 with no service_tier attribute (feature "On-demand Inference"), as one USD price dimension from 0 to Inf.
  • AWS lists this price in USD per 1K tokens (unit "1K tokens"); it is multiplied by 1,000 exactly, by decimal arithmetic, to give USD per 1M tokens.
  • AWS's on-demand price for this model with no service tier stated: the row carries no service_tier attribute, so it is not labelled as the standard or any other tier. Only this plain on-demand usagetype in us-east-1 is published: any flex, priority, batch, latency-optimized, custom-model or cross-region (global) variant, other AWS Regions, provisioned throughput and model customization are excluded. The offer file lists a cache read price for this model as a separate usagetype; cache write usagetypes are not published.

Input: $0.80 per 1M tokens · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-NovaPro-input-tokens.
  • Input tokens.

Output: $3.20 per 1M tokens · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-NovaPro-output-tokens.
  • Output tokens.

Cache read: $0.20 per 1M tokens · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-NovaPro-cache-read-input-token-count.
  • Cache read tokens (a "cache-read-input-token-count" usagetype). The offer file states no cache eligibility or cache lifetime for reads.

Price observed:

API availability: listed as available when checked (not a live service check)

Collection record

Primary source ↗ · provider documentation · Claims: bedrock-nova-pro: specifications · Retrieved · Checked · Reviewed · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card ("Models at a glance") for Nova Pro, Model ID amazon.nova-pro-v1:0. The offer's price list usagetype (USE1-NovaPro-input-tokens) is a price list name, not a model ID, so identity rests on the price list model attribute "Nova Pro" matching this card. Raw SHA-256 adc175e1d746ea5f313ad038331546c5e1610e9dfe703559934634dbaadcaea1, read once on 2026-11-07 under the #837 AWS facts-plus-link basis: "Context window: 300K tokens".
  • Shorthand recorded as stated (#504); not converted to an exact token count. The value is AWS's statement for its Bedrock listing; another seller of the same model may state a different figure.
  • Specifications only: no pricing, availability, capability, lifecycle or release-date assertion.

Raw API evidence ↗ · provider API · Claims: bedrock-nova-pro: pricing · bedrock-nova-pro: availability · 122 more claims about 61 other catalog entries · Retrieved · Checked · Review due · Confidence: A

  • Raw SHA-256: f1ac6a175fb27b3ad80d2c5d8970be8c2e5f957906349e52d15ec11d43e97896; collector aws-bedrock-price-list-json-1.
  • AWS Price List Bulk API offer file for Amazon Bedrock (us-east-1): offer version 20261111013037, publicationDate 2026-11-11T01:30:37Z.
  • Reviewed on-demand token prices only: each curated model's exact usagetypes in us-east-1, with service_tier "standard" for standard-tier models and with no service_tier attribute for the reviewed untiered models, in US dollars as the offer file states them (hence confidence A). A hosted listing and pricing observation only: no checkpoint, developer, license, context-window, modality or release-date assertions, and the usagetype is not a documented Bedrock model ID.
  • No Anthropic Claude price is published from this offer file: prices are read only from curated usagetypes, and no curated entry is an Anthropic model. (When the collector was reviewed on 2026-10-29, the file listed only legacy Claude models, all without a service tier.)
  • Availability rests on AWS's own offer file listing an on-demand price for the model in us-east-1.

NVIDIA Nemotron 3 Super 120B A12B

Context window: 256K tokens

All rates, conditions and sources

Conditions of every rate below:

  • Listed for model "NVIDIA Nemotron 3 Super 120B A12B" (provider "Nvidia") in us-east-1 with service_tier "standard", as one USD price dimension from 0 to Inf.
  • AWS lists this price in USD per 1K tokens (unit "1K tokens"); it is multiplied by 1,000 exactly, by decimal arithmetic, to give USD per 1M tokens.
  • Only the standard on-demand tier in us-east-1 is published: flex, priority, batch and global cross-region tiers, other AWS Regions and provisioned throughput are excluded. The offer file lists no standard-tier cache price for this model; missing cache prices are unknown, not zero.

Input: $0.15 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-nvidia.nemotron-super-3-120b-mantle-input-tokens-standard.
  • Input tokens.

Output: $0.65 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-nvidia.nemotron-super-3-120b-mantle-output-tokens-standard.
  • Output tokens.

Price observed:

API availability: listed as available when checked (not a live service check)

Collection record

Primary source ↗ · provider documentation · Claims: bedrock-nemotron-super-3-120b: specifications · Retrieved · Checked · Reviewed · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card ("Models at a glance") for NVIDIA Nemotron 3 Super 120B, Model ID nvidia.nemotron-super-3-120b (the offer's price list usagetype stem), raw SHA-256 45ff92d815a1ff5c59016bf7fec9acbe760156784159de46c875025f64f984a3, read once on 2026-10-31 under the #837 AWS facts-plus-link basis: "Context window: 256K tokens".
  • Shorthand recorded as stated (#504); not converted to an exact token count. The value is AWS's statement for its Bedrock listing; another seller of the same model may state a different figure.
  • Specifications only: no pricing, availability, capability, lifecycle or release-date assertion.

Raw API evidence ↗ · provider API · Claims: bedrock-nemotron-super-3-120b: pricing · bedrock-nemotron-super-3-120b: availability · 122 more claims about 61 other catalog entries · Retrieved · Checked · Review due · Confidence: A

  • Raw SHA-256: f1ac6a175fb27b3ad80d2c5d8970be8c2e5f957906349e52d15ec11d43e97896; collector aws-bedrock-price-list-json-1.
  • AWS Price List Bulk API offer file for Amazon Bedrock (us-east-1): offer version 20261111013037, publicationDate 2026-11-11T01:30:37Z.
  • Reviewed on-demand token prices only: each curated model's exact usagetypes in us-east-1, with service_tier "standard" for standard-tier models and with no service_tier attribute for the reviewed untiered models, in US dollars as the offer file states them (hence confidence A). A hosted listing and pricing observation only: no checkpoint, developer, license, context-window, modality or release-date assertions, and the usagetype is not a documented Bedrock model ID.
  • No Anthropic Claude price is published from this offer file: prices are read only from curated usagetypes, and no curated entry is an Anthropic model. (When the collector was reviewed on 2026-10-29, the file listed only legacy Claude models, all without a service tier.)
  • Availability rests on AWS's own offer file listing an on-demand price for the model in us-east-1.

NVIDIA Nemotron Nano 2

Context window: 128K tokens

Deprecated by provider · still listed · price checked Nov 12, 2026 (deprecation notice ↗)

All rates, conditions and sources

Conditions of every rate below:

  • Listed for model "NVIDIA Nemotron Nano 2" (provider "Nvidia") in us-east-1 with service_tier "standard", as one USD price dimension from 0 to Inf.
  • AWS lists this price in USD per 1K tokens (unit "1K tokens"); it is multiplied by 1,000 exactly, by decimal arithmetic, to give USD per 1M tokens.
  • Only the standard on-demand tier in us-east-1 is published: flex, priority, batch and global cross-region tiers, other AWS Regions and provisioned throughput are excluded. The offer file lists no standard-tier cache price for this model; missing cache prices are unknown, not zero.

Input: $0.06 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-nvidia.nemotron-nano-9b-v2-mantle-input-tokens-standard.
  • Input tokens.

Output: $0.23 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-nvidia.nemotron-nano-9b-v2-mantle-output-tokens-standard.
  • Output tokens.

Price observed:

API availability: listed as available when checked (not a live service check)

Collection record

Primary source ↗ · provider documentation · Claims: bedrock-nemotron-nano-9b-v2: specifications · Retrieved · Checked · Reviewed · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card ("Models at a glance") for NVIDIA Nemotron Nano 9B v2, Model ID nvidia.nemotron-nano-9b-v2 (the offer's price list usagetype stem), raw SHA-256 2dd0b9ab7e92a440180640908f51f8779832397d477a74c6194b8a8029faae73, read once on 2026-10-31 under the #837 AWS facts-plus-link basis: "Context window: 128K tokens".
  • Shorthand recorded as stated (#504); not converted to an exact token count. The value is AWS's statement for its Bedrock listing; another seller of the same model may state a different figure.
  • Specifications only: no pricing, availability, capability, lifecycle or release-date assertion.

Primary source ↗ · provider documentation · Claims: bedrock-nemotron-nano-9b-v2: availability · Retrieved · Checked · Reviewed · Review due · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card for NVIDIA Nemotron Nano 9B v2, Model ID nvidia.nemotron-nano-9b-v2, raw SHA-256 2dd0b9ab7e92a440180640908f51f8779832397d477a74c6194b8a8029faae73, read once on 2026-10-31 under the #837 AWS facts-plus-link basis: "Model lifecycle: Legacy"; "Model EOL date: March 30, 2027"; "Model lifecycle policy: Model lifecycle (For Models Launched Prior to Sept 7 2026)".
  • Amazon Bedrock "Model lifecycle (Legacy)" page (https://docs.aws.amazon.com/bedrock/latest/userguide/model-lifecycle-legacy.html), read once at 2026-10-31T16:36:03Z, raw SHA-256 c33d9d25d7da7f47c098aa864830361de770224f38a790070bada021dac56850: its Legacy table lists NVIDIA Nemotron Nano 9B v2 (nvidia.nemotron-nano-9b-v2) with Legacy date September 30, 2026, EOL date March 30, 2027 and public extended access start date "To be announced". The page states: "New customers can't use Legacy models and existing customers may lose access to Legacy models after 15 days of inactivity."
  • That table row's Regions include us-east-1, the region this offer is published for.
  • Availability stays available: the us-east-1 price list still lists an on-demand price, and the page says "you can continue to use a Legacy model during this period" (the Legacy period before the EOL date), though not as a new customer. The EOL date is in the future, so no retirement is recorded; reviewAfter is the EOL date, a re-review date, not a retirement.
  • Reviewed by Claude (owner delegation 2026-09-27), #909.

Raw API evidence ↗ · provider API · Claims: bedrock-nemotron-nano-9b-v2: pricing · bedrock-nemotron-nano-9b-v2: availability · 122 more claims about 61 other catalog entries · Retrieved · Checked · Review due · Confidence: A

  • Raw SHA-256: f1ac6a175fb27b3ad80d2c5d8970be8c2e5f957906349e52d15ec11d43e97896; collector aws-bedrock-price-list-json-1.
  • AWS Price List Bulk API offer file for Amazon Bedrock (us-east-1): offer version 20261111013037, publicationDate 2026-11-11T01:30:37Z.
  • Reviewed on-demand token prices only: each curated model's exact usagetypes in us-east-1, with service_tier "standard" for standard-tier models and with no service_tier attribute for the reviewed untiered models, in US dollars as the offer file states them (hence confidence A). A hosted listing and pricing observation only: no checkpoint, developer, license, context-window, modality or release-date assertions, and the usagetype is not a documented Bedrock model ID.
  • No Anthropic Claude price is published from this offer file: prices are read only from curated usagetypes, and no curated entry is an Anthropic model. (When the collector was reviewed on 2026-10-29, the file listed only legacy Claude models, all without a service tier.)
  • Availability rests on AWS's own offer file listing an on-demand price for the model in us-east-1.

NVIDIA Nemotron Nano 2 VL

Context window: 128K tokens

Deprecated by provider · still listed · price checked Nov 12, 2026 (deprecation notice ↗)

All rates, conditions and sources

Conditions of every rate below:

  • Listed for model "NVIDIA Nemotron Nano 2 VL" (provider "Nvidia") in us-east-1 with service_tier "standard", as one USD price dimension from 0 to Inf.
  • AWS lists this price in USD per 1K tokens (unit "1K tokens"); it is multiplied by 1,000 exactly, by decimal arithmetic, to give USD per 1M tokens.
  • Only the standard on-demand tier in us-east-1 is published: flex, priority, batch and global cross-region tiers, other AWS Regions and provisioned throughput are excluded. The offer file lists no standard-tier cache price for this model; missing cache prices are unknown, not zero.

Input: $0.20 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-nvidia.nemotron-nano-12b-v2-mantle-input-tokens-standard.
  • Input tokens.

Output: $0.60 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-nvidia.nemotron-nano-12b-v2-mantle-output-tokens-standard.
  • Output tokens.

Price observed:

API availability: listed as available when checked (not a live service check)

Collection record

Primary source ↗ · provider documentation · Claims: bedrock-nemotron-nano-12b-v2: specifications · Retrieved · Checked · Reviewed · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card ("Models at a glance") for NVIDIA Nemotron Nano 12B v2 VL BF16, Model ID nvidia.nemotron-nano-12b-v2 (the offer's price list usagetype stem), raw SHA-256 2e2bb2c5cd7528184995217fd83939989c46fdfa374e0e5f9f913e47a42b539f, read once on 2026-10-31 under the #837 AWS facts-plus-link basis: "Context window: 128K tokens".
  • Shorthand recorded as stated (#504); not converted to an exact token count. The value is AWS's statement for its Bedrock listing; another seller of the same model may state a different figure.
  • Specifications only: no pricing, availability, capability, lifecycle or release-date assertion.

Primary source ↗ · provider documentation · Claims: bedrock-nemotron-nano-12b-v2: availability · Retrieved · Checked · Reviewed · Review due · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card for NVIDIA Nemotron Nano 12B v2 VL BF16, Model ID nvidia.nemotron-nano-12b-v2, raw SHA-256 2e2bb2c5cd7528184995217fd83939989c46fdfa374e0e5f9f913e47a42b539f, read once on 2026-10-31 under the #837 AWS facts-plus-link basis: "Model lifecycle: Legacy"; "Model EOL date: March 30, 2027"; "Model lifecycle policy: Model lifecycle (For Models Launched Prior to Sept 7 2026)".
  • Amazon Bedrock "Model lifecycle (Legacy)" page (https://docs.aws.amazon.com/bedrock/latest/userguide/model-lifecycle-legacy.html), read once at 2026-10-31T16:36:03Z, raw SHA-256 c33d9d25d7da7f47c098aa864830361de770224f38a790070bada021dac56850: its Legacy table lists NVIDIA Nemotron Nano 12B v2 VL BF16 (nvidia.nemotron-nano-12b-v2) with Legacy date September 30, 2026, EOL date March 30, 2027 and public extended access start date "To be announced". The page states: "New customers can't use Legacy models and existing customers may lose access to Legacy models after 15 days of inactivity."
  • That table row's Regions include us-east-1, the region this offer is published for.
  • Availability stays available: the us-east-1 price list still lists an on-demand price, and the page says "you can continue to use a Legacy model during this period" (the Legacy period before the EOL date), though not as a new customer. The EOL date is in the future, so no retirement is recorded; reviewAfter is the EOL date, a re-review date, not a retirement.
  • Reviewed by Claude (owner delegation 2026-09-27), #909.

Raw API evidence ↗ · provider API · Claims: bedrock-nemotron-nano-12b-v2: pricing · bedrock-nemotron-nano-12b-v2: availability · 122 more claims about 61 other catalog entries · Retrieved · Checked · Review due · Confidence: A

  • Raw SHA-256: f1ac6a175fb27b3ad80d2c5d8970be8c2e5f957906349e52d15ec11d43e97896; collector aws-bedrock-price-list-json-1.
  • AWS Price List Bulk API offer file for Amazon Bedrock (us-east-1): offer version 20261111013037, publicationDate 2026-11-11T01:30:37Z.
  • Reviewed on-demand token prices only: each curated model's exact usagetypes in us-east-1, with service_tier "standard" for standard-tier models and with no service_tier attribute for the reviewed untiered models, in US dollars as the offer file states them (hence confidence A). A hosted listing and pricing observation only: no checkpoint, developer, license, context-window, modality or release-date assertions, and the usagetype is not a documented Bedrock model ID.
  • No Anthropic Claude price is published from this offer file: prices are read only from curated usagetypes, and no curated entry is an Anthropic model. (When the collector was reviewed on 2026-10-29, the file listed only legacy Claude models, all without a service tier.)
  • Availability rests on AWS's own offer file listing an on-demand price for the model in us-east-1.

Pixtral Large 25.02

Context window: 128K tokens

All rates, conditions and sources

Conditions of every rate below:

  • Listed for model "Pixtral Large 25.02" (provider "Mistral") in us-east-1 with no service_tier attribute (feature "On-demand Inference"), as one USD price dimension from 0 to Inf.
  • AWS lists this price in USD per 1K tokens (unit "1K tokens"); it is multiplied by 1,000 exactly, by decimal arithmetic, to give USD per 1M tokens.
  • AWS's on-demand price for this model with no service tier stated: the row carries no service_tier attribute, so it is not labelled as the standard or any other tier. Only this plain on-demand usagetype in us-east-1 is published: any flex, priority, batch, latency-optimized, custom-model or cross-region (global) variant, other AWS Regions, provisioned throughput and model customization are excluded. The offer file lists no cache price for this model without a service tier; missing cache prices are unknown, not zero.

Input: $2.00 per 1M tokens · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-PixtralLarge2502-input-tokens.
  • Input tokens.

Output: $6.00 per 1M tokens · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-PixtralLarge2502-output-tokens.
  • Output tokens.

Price observed:

API availability: listed as available when checked (not a live service check)

Collection record

Primary source ↗ · provider documentation · Claims: bedrock-pixtral-large-2502: specifications · Retrieved · Checked · Reviewed · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card ("Models at a glance") for Pixtral Large, Model ID mistral.pixtral-large-2502-v1:0. The offer's price list usagetype (USE1-PixtralLarge2502-input-tokens) is a price list name, not a model ID, so identity rests on the price list model attribute "Pixtral Large 25.02" matching this card. The attribute adds the version "25.02", which the card's Model ID carries (pixtral-large-2502). Raw SHA-256 49a8538b407a46ad29f5c2abc4b5ab741ba16bf4c235a43ac8d8166a59fb4781, read once on 2026-11-07 under the #837 AWS facts-plus-link basis: "Context window: 128K tokens".
  • Shorthand recorded as stated (#504); not converted to an exact token count. The value is AWS's statement for its Bedrock listing; another seller of the same model may state a different figure.
  • Specifications only: no pricing, availability, capability, lifecycle or release-date assertion.

Raw API evidence ↗ · provider API · Claims: bedrock-pixtral-large-2502: pricing · bedrock-pixtral-large-2502: availability · 122 more claims about 61 other catalog entries · Retrieved · Checked · Review due · Confidence: A

  • Raw SHA-256: f1ac6a175fb27b3ad80d2c5d8970be8c2e5f957906349e52d15ec11d43e97896; collector aws-bedrock-price-list-json-1.
  • AWS Price List Bulk API offer file for Amazon Bedrock (us-east-1): offer version 20261111013037, publicationDate 2026-11-11T01:30:37Z.
  • Reviewed on-demand token prices only: each curated model's exact usagetypes in us-east-1, with service_tier "standard" for standard-tier models and with no service_tier attribute for the reviewed untiered models, in US dollars as the offer file states them (hence confidence A). A hosted listing and pricing observation only: no checkpoint, developer, license, context-window, modality or release-date assertions, and the usagetype is not a documented Bedrock model ID.
  • No Anthropic Claude price is published from this offer file: prices are read only from curated usagetypes, and no curated entry is an Anthropic model. (When the collector was reviewed on 2026-10-29, the file listed only legacy Claude models, all without a service tier.)
  • Availability rests on AWS's own offer file listing an on-demand price for the model in us-east-1.

Qwen3 235B A22B 2507

Context window: 256K tokens

All rates, conditions and sources

Conditions of every rate below:

  • Listed for model "Qwen3 235B A22B 2507" (provider "Qwen") in us-east-1 with service_tier "standard", as one USD price dimension from 0 to Inf.
  • AWS lists this price in USD per 1K tokens (unit "1K tokens"); it is multiplied by 1,000 exactly, by decimal arithmetic, to give USD per 1M tokens.
  • Only the standard on-demand tier in us-east-1 is published: flex, priority, batch and global cross-region tiers, other AWS Regions and provisioned throughput are excluded. The offer file lists no standard-tier cache price for this model; missing cache prices are unknown, not zero.

Input: $0.22 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-qwen.qwen3-235b-a22b-2507-mantle-input-tokens-standard.
  • Input tokens.

Output: $0.88 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-qwen.qwen3-235b-a22b-2507-mantle-output-tokens-standard.
  • Output tokens.

Price observed:

API availability: listed as available when checked (not a live service check)

Collection record

Primary source ↗ · provider documentation · Claims: bedrock-qwen3-235b-a22b-2507: specifications · Retrieved · Checked · Reviewed · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card ("Models at a glance") for Qwen3 235B A22B 2507, Model ID qwen.qwen3-235b-a22b-2507 (the offer's price list usagetype stem), raw SHA-256 640f2b1d9a69b77e90086e414d7c20c755b24cf81d15f22a216222184abf9735, read once on 2026-10-31 under the #837 AWS facts-plus-link basis: "Context window: 256K tokens".
  • Shorthand recorded as stated (#504); not converted to an exact token count. The value is AWS's statement for its Bedrock listing; another seller of the same model may state a different figure.
  • Specifications only: no pricing, availability, capability, lifecycle or release-date assertion.

Raw API evidence ↗ · provider API · Claims: bedrock-qwen3-235b-a22b-2507: pricing · bedrock-qwen3-235b-a22b-2507: availability · 122 more claims about 61 other catalog entries · Retrieved · Checked · Review due · Confidence: A

  • Raw SHA-256: f1ac6a175fb27b3ad80d2c5d8970be8c2e5f957906349e52d15ec11d43e97896; collector aws-bedrock-price-list-json-1.
  • AWS Price List Bulk API offer file for Amazon Bedrock (us-east-1): offer version 20261111013037, publicationDate 2026-11-11T01:30:37Z.
  • Reviewed on-demand token prices only: each curated model's exact usagetypes in us-east-1, with service_tier "standard" for standard-tier models and with no service_tier attribute for the reviewed untiered models, in US dollars as the offer file states them (hence confidence A). A hosted listing and pricing observation only: no checkpoint, developer, license, context-window, modality or release-date assertions, and the usagetype is not a documented Bedrock model ID.
  • No Anthropic Claude price is published from this offer file: prices are read only from curated usagetypes, and no curated entry is an Anthropic model. (When the collector was reviewed on 2026-10-29, the file listed only legacy Claude models, all without a service tier.)
  • Availability rests on AWS's own offer file listing an on-demand price for the model in us-east-1.

Qwen3 32B

Context window: 32K tokens

All rates, conditions and sources

Conditions of every rate below:

  • Listed for model "Qwen3 32B" (provider "Qwen") in us-east-1 with service_tier "standard", as one USD price dimension from 0 to Inf.
  • AWS lists this price in USD per 1K tokens (unit "1K tokens"); it is multiplied by 1,000 exactly, by decimal arithmetic, to give USD per 1M tokens.
  • Only the standard on-demand tier in us-east-1 is published: flex, priority, batch and global cross-region tiers, other AWS Regions and provisioned throughput are excluded. The offer file lists no standard-tier cache price for this model; missing cache prices are unknown, not zero.

Input: $0.15 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-qwen.qwen3-32b-mantle-input-tokens-standard.
  • Input tokens.

Output: $0.60 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-qwen.qwen3-32b-mantle-output-tokens-standard.
  • Output tokens.

Price observed:

API availability: listed as available when checked (not a live service check)

Collection record

Primary source ↗ · provider documentation · Claims: bedrock-qwen3-32b: specifications · Retrieved · Checked · Reviewed · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card ("Models at a glance") for Qwen3 32B, Model ID qwen.qwen3-32b (the offer's price list usagetype stem), raw SHA-256 6322edcbdbf39ae664967b3c9e76d67427ed46c8672862526b051c883d73276f, read once on 2026-10-31 under the #837 AWS facts-plus-link basis: "Context window: 32K tokens".
  • Shorthand recorded as stated (#504); not converted to an exact token count. The value is AWS's statement for its Bedrock listing; another seller of the same model may state a different figure.
  • Specifications only: no pricing, availability, capability, lifecycle or release-date assertion.

Raw API evidence ↗ · provider API · Claims: bedrock-qwen3-32b: pricing · bedrock-qwen3-32b: availability · 122 more claims about 61 other catalog entries · Retrieved · Checked · Review due · Confidence: A

  • Raw SHA-256: f1ac6a175fb27b3ad80d2c5d8970be8c2e5f957906349e52d15ec11d43e97896; collector aws-bedrock-price-list-json-1.
  • AWS Price List Bulk API offer file for Amazon Bedrock (us-east-1): offer version 20261111013037, publicationDate 2026-11-11T01:30:37Z.
  • Reviewed on-demand token prices only: each curated model's exact usagetypes in us-east-1, with service_tier "standard" for standard-tier models and with no service_tier attribute for the reviewed untiered models, in US dollars as the offer file states them (hence confidence A). A hosted listing and pricing observation only: no checkpoint, developer, license, context-window, modality or release-date assertions, and the usagetype is not a documented Bedrock model ID.
  • No Anthropic Claude price is published from this offer file: prices are read only from curated usagetypes, and no curated entry is an Anthropic model. (When the collector was reviewed on 2026-10-29, the file listed only legacy Claude models, all without a service tier.)
  • Availability rests on AWS's own offer file listing an on-demand price for the model in us-east-1.

Qwen3 Coder 30B A3B

Context window: 256K tokens

Deprecated by provider · still listed · price checked Nov 12, 2026 (deprecation notice ↗)

All rates, conditions and sources

Conditions of every rate below:

  • Listed for model "Qwen3 Coder 30B A3B" (provider "Qwen") in us-east-1 with service_tier "standard", as one USD price dimension from 0 to Inf.
  • AWS lists this price in USD per 1K tokens (unit "1K tokens"); it is multiplied by 1,000 exactly, by decimal arithmetic, to give USD per 1M tokens.
  • Only the standard on-demand tier in us-east-1 is published: flex, priority, batch and global cross-region tiers, other AWS Regions and provisioned throughput are excluded. The offer file lists no standard-tier cache price for this model; missing cache prices are unknown, not zero.

Input: $0.15 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-qwen.qwen3-coder-30b-a3b-instruct-mantle-input-tokens-standard.
  • Input tokens.

Output: $0.60 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-qwen.qwen3-coder-30b-a3b-instruct-mantle-output-tokens-standard.
  • Output tokens.

Price observed:

API availability: listed as available when checked (not a live service check)

Collection record

Primary source ↗ · provider documentation · Claims: bedrock-qwen3-coder-30b-a3b-instruct: specifications · Retrieved · Checked · Reviewed · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card ("Models at a glance") for Qwen3-Coder-30B-A3B-Instruct, Model ID qwen.qwen3-coder-30b-a3b-instruct (the offer's price list usagetype stem), raw SHA-256 10feba5dcfc3b02e9d37ad380d7c31cb9765ad38f8f647a0bb3828883902caa5, read once on 2026-10-31 under the #837 AWS facts-plus-link basis: "Context window: 256K tokens".
  • Shorthand recorded as stated (#504); not converted to an exact token count. The value is AWS's statement for its Bedrock listing; another seller of the same model may state a different figure.
  • Specifications only: no pricing, availability, capability, lifecycle or release-date assertion.

Primary source ↗ · provider documentation · Claims: bedrock-qwen3-coder-30b-a3b-instruct: availability · Retrieved · Checked · Reviewed · Review due · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card for Qwen3-Coder-30B-A3B-Instruct, Model ID qwen.qwen3-coder-30b-a3b-instruct, raw SHA-256 10feba5dcfc3b02e9d37ad380d7c31cb9765ad38f8f647a0bb3828883902caa5, read once on 2026-10-31 under the #837 AWS facts-plus-link basis: "Model lifecycle: Legacy"; "Model EOL date: March 30, 2027"; "Model lifecycle policy: Model lifecycle (For Models Launched Prior to Sept 7 2026)".
  • Amazon Bedrock "Model lifecycle (Legacy)" page (https://docs.aws.amazon.com/bedrock/latest/userguide/model-lifecycle-legacy.html), read once at 2026-10-31T16:36:03Z, raw SHA-256 c33d9d25d7da7f47c098aa864830361de770224f38a790070bada021dac56850: its Legacy table lists Qwen3-Coder-30B-A3B-Instruct (qwen.qwen3-coder-30b-a3b-v1:0 (bedrock-runtime), qwen.qwen3-coder-30b-a3b-instruct (bedrock-mantle)) with Legacy date September 30, 2026, EOL date March 30, 2027 and public extended access start date "To be announced". The page states: "New customers can't use Legacy models and existing customers may lose access to Legacy models after 15 days of inactivity."
  • That table row's Regions include us-east-1, the region this offer is published for.
  • Availability stays available: the us-east-1 price list still lists an on-demand price, and the page says "you can continue to use a Legacy model during this period" (the Legacy period before the EOL date), though not as a new customer. The EOL date is in the future, so no retirement is recorded; reviewAfter is the EOL date, a re-review date, not a retirement.
  • Reviewed by Claude (owner delegation 2026-09-27), #909.

Raw API evidence ↗ · provider API · Claims: bedrock-qwen3-coder-30b-a3b-instruct: pricing · bedrock-qwen3-coder-30b-a3b-instruct: availability · 122 more claims about 61 other catalog entries · Retrieved · Checked · Review due · Confidence: A

  • Raw SHA-256: f1ac6a175fb27b3ad80d2c5d8970be8c2e5f957906349e52d15ec11d43e97896; collector aws-bedrock-price-list-json-1.
  • AWS Price List Bulk API offer file for Amazon Bedrock (us-east-1): offer version 20261111013037, publicationDate 2026-11-11T01:30:37Z.
  • Reviewed on-demand token prices only: each curated model's exact usagetypes in us-east-1, with service_tier "standard" for standard-tier models and with no service_tier attribute for the reviewed untiered models, in US dollars as the offer file states them (hence confidence A). A hosted listing and pricing observation only: no checkpoint, developer, license, context-window, modality or release-date assertions, and the usagetype is not a documented Bedrock model ID.
  • No Anthropic Claude price is published from this offer file: prices are read only from curated usagetypes, and no curated entry is an Anthropic model. (When the collector was reviewed on 2026-10-29, the file listed only legacy Claude models, all without a service tier.)
  • Availability rests on AWS's own offer file listing an on-demand price for the model in us-east-1.

Qwen3 Coder 480B A35B

Context window: 256K tokens

Deprecated by provider · still listed · price checked Nov 12, 2026 (deprecation notice ↗)

All rates, conditions and sources

Conditions of every rate below:

  • Listed for model "Qwen3 Coder 480B A35B" (provider "Qwen") in us-east-1 with service_tier "standard", as one USD price dimension from 0 to Inf.
  • AWS lists this price in USD per 1K tokens (unit "1K tokens"); it is multiplied by 1,000 exactly, by decimal arithmetic, to give USD per 1M tokens.
  • Only the standard on-demand tier in us-east-1 is published: flex, priority, batch and global cross-region tiers, other AWS Regions and provisioned throughput are excluded. The offer file lists no standard-tier cache price for this model; missing cache prices are unknown, not zero.

Input: $0.45 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-qwen.qwen3-coder-480b-a35b-instruct-mantle-input-tokens-standard.
  • Input tokens.

Output: $1.80 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-qwen.qwen3-coder-480b-a35b-instruct-mantle-output-tokens-standard.
  • Output tokens.

Price observed:

API availability: listed as available when checked (not a live service check)

Collection record

Primary source ↗ · provider documentation · Claims: bedrock-qwen3-coder-480b-a35b-instruct: specifications · Retrieved · Checked · Reviewed · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card ("Models at a glance") for Qwen3 Coder 480B A35B Instruct, Model ID qwen.qwen3-coder-480b-a35b-instruct (the offer's price list usagetype stem), raw SHA-256 3151bd67b3ed1915e4a4fcdc654f241f174c9e8be408453b3350381dd3cc3169, read once on 2026-10-31 under the #837 AWS facts-plus-link basis: "Context window: 256K tokens".
  • Shorthand recorded as stated (#504); not converted to an exact token count. The value is AWS's statement for its Bedrock listing; another seller of the same model may state a different figure.
  • Specifications only: no pricing, availability, capability, lifecycle or release-date assertion.

Primary source ↗ · provider documentation · Claims: bedrock-qwen3-coder-480b-a35b-instruct: availability · Retrieved · Checked · Reviewed · Review due · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card for Qwen3 Coder 480B A35B Instruct, Model ID qwen.qwen3-coder-480b-a35b-instruct, raw SHA-256 3151bd67b3ed1915e4a4fcdc654f241f174c9e8be408453b3350381dd3cc3169, read once on 2026-10-31 under the #837 AWS facts-plus-link basis: "Model lifecycle: Legacy"; "Model EOL date: March 30, 2027"; "Model lifecycle policy: Model lifecycle (For Models Launched Prior to Sept 7 2026)".
  • Amazon Bedrock "Model lifecycle (Legacy)" page (https://docs.aws.amazon.com/bedrock/latest/userguide/model-lifecycle-legacy.html), read once at 2026-10-31T16:36:03Z, raw SHA-256 c33d9d25d7da7f47c098aa864830361de770224f38a790070bada021dac56850: its Legacy table lists Qwen3 Coder 480B A35B Instruct (qwen.qwen3-coder-480b-a35b-v1:0 (bedrock-runtime), qwen.qwen3-coder-480b-a35b-instruct (bedrock-mantle)) with Legacy date September 30, 2026, EOL date March 30, 2027 and public extended access start date "To be announced". The page states: "New customers can't use Legacy models and existing customers may lose access to Legacy models after 15 days of inactivity."
  • That table row's Regions (ap-northeast-1, ap-south-1, ap-southeast-2, ap-southeast-3, eu-north-1, eu-west-2, us-east-2, us-west-2) does not list us-east-1, the only region this offer is published for. The card's own "Model lifecycle: Legacy" field is a statement about the model, so the us-east-1 listing is recorded as deprecated, with this gap noted; whether AWS applies the Legacy state to us-east-1 is not stated.
  • Availability stays available: the us-east-1 price list still lists an on-demand price, and the page says "you can continue to use a Legacy model during this period" (the Legacy period before the EOL date), though not as a new customer. The EOL date is in the future, so no retirement is recorded; reviewAfter is the EOL date, a re-review date, not a retirement.
  • Reviewed by Claude (owner delegation 2026-09-27), #909.

Raw API evidence ↗ · provider API · Claims: bedrock-qwen3-coder-480b-a35b-instruct: pricing · bedrock-qwen3-coder-480b-a35b-instruct: availability · 122 more claims about 61 other catalog entries · Retrieved · Checked · Review due · Confidence: A

  • Raw SHA-256: f1ac6a175fb27b3ad80d2c5d8970be8c2e5f957906349e52d15ec11d43e97896; collector aws-bedrock-price-list-json-1.
  • AWS Price List Bulk API offer file for Amazon Bedrock (us-east-1): offer version 20261111013037, publicationDate 2026-11-11T01:30:37Z.
  • Reviewed on-demand token prices only: each curated model's exact usagetypes in us-east-1, with service_tier "standard" for standard-tier models and with no service_tier attribute for the reviewed untiered models, in US dollars as the offer file states them (hence confidence A). A hosted listing and pricing observation only: no checkpoint, developer, license, context-window, modality or release-date assertions, and the usagetype is not a documented Bedrock model ID.
  • No Anthropic Claude price is published from this offer file: prices are read only from curated usagetypes, and no curated entry is an Anthropic model. (When the collector was reviewed on 2026-10-29, the file listed only legacy Claude models, all without a service tier.)
  • Availability rests on AWS's own offer file listing an on-demand price for the model in us-east-1.

Qwen3 Coder Next

Context window: 256K tokens

All rates, conditions and sources

Conditions of every rate below:

  • Listed for model "Qwen3 Coder Next" (provider "Qwen") in us-east-1 with service_tier "standard", as one USD price dimension from 0 to Inf.
  • AWS lists this price in USD per 1K tokens (unit "1K tokens"); it is multiplied by 1,000 exactly, by decimal arithmetic, to give USD per 1M tokens.
  • Only the standard on-demand tier in us-east-1 is published: flex, priority, batch and global cross-region tiers, other AWS Regions and provisioned throughput are excluded. The offer file lists no standard-tier cache price for this model; missing cache prices are unknown, not zero.

Input: $0.50 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-qwen.qwen3-coder-next-mantle-input-tokens-standard.
  • Input tokens.

Output: $1.20 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-qwen.qwen3-coder-next-mantle-output-tokens-standard.
  • Output tokens.

Price observed:

API availability: listed as available when checked (not a live service check)

Collection record

Primary source ↗ · provider documentation · Claims: bedrock-qwen3-coder-next: specifications · Retrieved · Checked · Reviewed · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card ("Models at a glance") for Qwen3 Coder Next, Model ID qwen.qwen3-coder-next (the offer's price list usagetype stem), raw SHA-256 d8dfabaa05f47564fd7df3c0aa3a3fa042ba49c4a6fee0fdf25583996084bfe0, read once on 2026-10-31 under the #837 AWS facts-plus-link basis: "Context window: 256K tokens".
  • Shorthand recorded as stated (#504); not converted to an exact token count. The value is AWS's statement for its Bedrock listing; another seller of the same model may state a different figure.
  • Specifications only: no pricing, availability, capability, lifecycle or release-date assertion.

Raw API evidence ↗ · provider API · Claims: bedrock-qwen3-coder-next: pricing · bedrock-qwen3-coder-next: availability · 122 more claims about 61 other catalog entries · Retrieved · Checked · Review due · Confidence: A

  • Raw SHA-256: f1ac6a175fb27b3ad80d2c5d8970be8c2e5f957906349e52d15ec11d43e97896; collector aws-bedrock-price-list-json-1.
  • AWS Price List Bulk API offer file for Amazon Bedrock (us-east-1): offer version 20261111013037, publicationDate 2026-11-11T01:30:37Z.
  • Reviewed on-demand token prices only: each curated model's exact usagetypes in us-east-1, with service_tier "standard" for standard-tier models and with no service_tier attribute for the reviewed untiered models, in US dollars as the offer file states them (hence confidence A). A hosted listing and pricing observation only: no checkpoint, developer, license, context-window, modality or release-date assertions, and the usagetype is not a documented Bedrock model ID.
  • No Anthropic Claude price is published from this offer file: prices are read only from curated usagetypes, and no curated entry is an Anthropic model. (When the collector was reviewed on 2026-10-29, the file listed only legacy Claude models, all without a service tier.)
  • Availability rests on AWS's own offer file listing an on-demand price for the model in us-east-1.

Qwen3 Next 80B A3B

Context window: 256K tokens

All rates, conditions and sources

Conditions of every rate below:

  • Listed for model "Qwen3 Next 80B A3B" (provider "Qwen") in us-east-1 with service_tier "standard", as one USD price dimension from 0 to Inf.
  • AWS lists this price in USD per 1K tokens (unit "1K tokens"); it is multiplied by 1,000 exactly, by decimal arithmetic, to give USD per 1M tokens.
  • Only the standard on-demand tier in us-east-1 is published: flex, priority, batch and global cross-region tiers, other AWS Regions and provisioned throughput are excluded. The offer file lists no standard-tier cache price for this model; missing cache prices are unknown, not zero.

Input: $0.14 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-qwen.qwen3-next-80b-a3b-instruct-mantle-input-tokens-standard.
  • Input tokens.

Output: $1.20 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-qwen.qwen3-next-80b-a3b-instruct-mantle-output-tokens-standard.
  • Output tokens.

Price observed:

API availability: listed as available when checked (not a live service check)

Collection record

Primary source ↗ · provider documentation · Claims: bedrock-qwen3-next-80b-a3b-instruct: specifications · Retrieved · Checked · Reviewed · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card ("Models at a glance") for Qwen3 Next 80B A3B, Model ID qwen.qwen3-next-80b-a3b-instruct (the offer's price list usagetype stem), raw SHA-256 0b9b0468a8352278b90fca8c9bcf80719c845deb07ff311499269b29998433b2, read once on 2026-10-31 under the #837 AWS facts-plus-link basis: "Context window: 256K tokens".
  • Shorthand recorded as stated (#504); not converted to an exact token count. The value is AWS's statement for its Bedrock listing; another seller of the same model may state a different figure.
  • Specifications only: no pricing, availability, capability, lifecycle or release-date assertion.

Raw API evidence ↗ · provider API · Claims: bedrock-qwen3-next-80b-a3b-instruct: pricing · bedrock-qwen3-next-80b-a3b-instruct: availability · 122 more claims about 61 other catalog entries · Retrieved · Checked · Review due · Confidence: A

  • Raw SHA-256: f1ac6a175fb27b3ad80d2c5d8970be8c2e5f957906349e52d15ec11d43e97896; collector aws-bedrock-price-list-json-1.
  • AWS Price List Bulk API offer file for Amazon Bedrock (us-east-1): offer version 20261111013037, publicationDate 2026-11-11T01:30:37Z.
  • Reviewed on-demand token prices only: each curated model's exact usagetypes in us-east-1, with service_tier "standard" for standard-tier models and with no service_tier attribute for the reviewed untiered models, in US dollars as the offer file states them (hence confidence A). A hosted listing and pricing observation only: no checkpoint, developer, license, context-window, modality or release-date assertions, and the usagetype is not a documented Bedrock model ID.
  • No Anthropic Claude price is published from this offer file: prices are read only from curated usagetypes, and no curated entry is an Anthropic model. (When the collector was reviewed on 2026-10-29, the file listed only legacy Claude models, all without a service tier.)
  • Availability rests on AWS's own offer file listing an on-demand price for the model in us-east-1.

Qwen3 VL 235B A22B

Context window: 256K tokens

All rates, conditions and sources

Conditions of every rate below:

  • Listed for model "Qwen3 VL 235B A22B" (provider "Qwen") in us-east-1 with service_tier "standard", as one USD price dimension from 0 to Inf.
  • AWS lists this price in USD per 1K tokens (unit "1K tokens"); it is multiplied by 1,000 exactly, by decimal arithmetic, to give USD per 1M tokens.
  • Only the standard on-demand tier in us-east-1 is published: flex, priority, batch and global cross-region tiers, other AWS Regions and provisioned throughput are excluded. The offer file lists no standard-tier cache price for this model; missing cache prices are unknown, not zero.

Input: $0.53 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-qwen.qwen3-vl-235b-a22b-instruct-mantle-input-tokens-standard.
  • Input tokens.

Output: $2.66 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-qwen.qwen3-vl-235b-a22b-instruct-mantle-output-tokens-standard.
  • Output tokens.

Price observed:

API availability: listed as available when checked (not a live service check)

Collection record

Primary source ↗ · provider documentation · Claims: bedrock-qwen3-vl-235b-a22b-instruct: specifications · Retrieved · Checked · Reviewed · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card ("Models at a glance") for Qwen3 VL 235B A22B, Model ID qwen.qwen3-vl-235b-a22b-instruct (the offer's price list usagetype stem), raw SHA-256 1a1370a18dd1df369e8f966f423ff2834091538e59312a627672c39d295fe557, read once on 2026-10-31 under the #837 AWS facts-plus-link basis: "Context window: 256K tokens".
  • Shorthand recorded as stated (#504); not converted to an exact token count. The value is AWS's statement for its Bedrock listing; another seller of the same model may state a different figure.
  • Specifications only: no pricing, availability, capability, lifecycle or release-date assertion.

Raw API evidence ↗ · provider API · Claims: bedrock-qwen3-vl-235b-a22b-instruct: pricing · bedrock-qwen3-vl-235b-a22b-instruct: availability · 122 more claims about 61 other catalog entries · Retrieved · Checked · Review due · Confidence: A

  • Raw SHA-256: f1ac6a175fb27b3ad80d2c5d8970be8c2e5f957906349e52d15ec11d43e97896; collector aws-bedrock-price-list-json-1.
  • AWS Price List Bulk API offer file for Amazon Bedrock (us-east-1): offer version 20261111013037, publicationDate 2026-11-11T01:30:37Z.
  • Reviewed on-demand token prices only: each curated model's exact usagetypes in us-east-1, with service_tier "standard" for standard-tier models and with no service_tier attribute for the reviewed untiered models, in US dollars as the offer file states them (hence confidence A). A hosted listing and pricing observation only: no checkpoint, developer, license, context-window, modality or release-date assertions, and the usagetype is not a documented Bedrock model ID.
  • No Anthropic Claude price is published from this offer file: prices are read only from curated usagetypes, and no curated entry is an Anthropic model. (When the collector was reviewed on 2026-10-29, the file listed only legacy Claude models, all without a service tier.)
  • Availability rests on AWS's own offer file listing an on-demand price for the model in us-east-1.

Voxtral Mini 1.0

Context window: 32K tokens

Deprecated by provider · still listed · price checked Nov 12, 2026 (deprecation notice ↗)

All rates, conditions and sources

Conditions of every rate below:

  • Listed for model "Voxtral Mini 1.0" (provider "Mistral") in us-east-1 with service_tier "standard", as one USD price dimension from 0 to Inf.
  • AWS lists this price in USD per 1K tokens (unit "1K tokens"); it is multiplied by 1,000 exactly, by decimal arithmetic, to give USD per 1M tokens.
  • Only the standard on-demand tier in us-east-1 is published: flex, priority, batch and global cross-region tiers, other AWS Regions and provisioned throughput are excluded. The offer file lists no standard-tier cache price for this model; missing cache prices are unknown, not zero.

Input: $0.04 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-mistral.voxtral-mini-3b-2507-mantle-input-tokens-standard.
  • Input tokens.

Output: $0.04 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-mistral.voxtral-mini-3b-2507-mantle-output-tokens-standard.
  • Output tokens.

Price observed:

API availability: listed as available when checked (not a live service check)

Collection record

Primary source ↗ · provider documentation · Claims: bedrock-voxtral-mini-3b-2507: specifications · Retrieved · Checked · Reviewed · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card ("Models at a glance") for Voxtral Mini 3B 2507, Model ID mistral.voxtral-mini-3b-2507 (the offer's price list usagetype stem), raw SHA-256 e944fd018f6e071d9eefa5d79d15300176d0f266e36e823a31d98c50adfb0cbc, read once on 2026-10-31 under the #837 AWS facts-plus-link basis: "Context window: 32K tokens".
  • Shorthand recorded as stated (#504); not converted to an exact token count. The value is AWS's statement for its Bedrock listing; another seller of the same model may state a different figure.
  • Specifications only: no pricing, availability, capability, lifecycle or release-date assertion.

Primary source ↗ · provider documentation · Claims: bedrock-voxtral-mini-3b-2507: availability · Retrieved · Checked · Reviewed · Review due · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card for Voxtral Mini 3B 2507, Model ID mistral.voxtral-mini-3b-2507, raw SHA-256 e944fd018f6e071d9eefa5d79d15300176d0f266e36e823a31d98c50adfb0cbc, read once on 2026-10-31 under the #837 AWS facts-plus-link basis: "Model lifecycle: Legacy"; "Model EOL date: March 30, 2027"; "Model lifecycle policy: Model lifecycle (For Models Launched Prior to Sept 7 2026)".
  • Amazon Bedrock "Model lifecycle (Legacy)" page (https://docs.aws.amazon.com/bedrock/latest/userguide/model-lifecycle-legacy.html), read once at 2026-10-31T16:36:03Z, raw SHA-256 c33d9d25d7da7f47c098aa864830361de770224f38a790070bada021dac56850: its Legacy table lists Voxtral Mini 3B 2507 (mistral.voxtral-mini-3b-2507) with Legacy date September 30, 2026, EOL date March 30, 2027 and public extended access start date "To be announced". The page states: "New customers can't use Legacy models and existing customers may lose access to Legacy models after 15 days of inactivity."
  • That table row's Regions include us-east-1, the region this offer is published for.
  • Availability stays available: the us-east-1 price list still lists an on-demand price, and the page says "you can continue to use a Legacy model during this period" (the Legacy period before the EOL date), though not as a new customer. The EOL date is in the future, so no retirement is recorded; reviewAfter is the EOL date, a re-review date, not a retirement.
  • Reviewed by Claude (owner delegation 2026-09-27), #909.

Raw API evidence ↗ · provider API · Claims: bedrock-voxtral-mini-3b-2507: pricing · bedrock-voxtral-mini-3b-2507: availability · 122 more claims about 61 other catalog entries · Retrieved · Checked · Review due · Confidence: A

  • Raw SHA-256: f1ac6a175fb27b3ad80d2c5d8970be8c2e5f957906349e52d15ec11d43e97896; collector aws-bedrock-price-list-json-1.
  • AWS Price List Bulk API offer file for Amazon Bedrock (us-east-1): offer version 20261111013037, publicationDate 2026-11-11T01:30:37Z.
  • Reviewed on-demand token prices only: each curated model's exact usagetypes in us-east-1, with service_tier "standard" for standard-tier models and with no service_tier attribute for the reviewed untiered models, in US dollars as the offer file states them (hence confidence A). A hosted listing and pricing observation only: no checkpoint, developer, license, context-window, modality or release-date assertions, and the usagetype is not a documented Bedrock model ID.
  • No Anthropic Claude price is published from this offer file: prices are read only from curated usagetypes, and no curated entry is an Anthropic model. (When the collector was reviewed on 2026-10-29, the file listed only legacy Claude models, all without a service tier.)
  • Availability rests on AWS's own offer file listing an on-demand price for the model in us-east-1.

Voxtral Small 1.0

Context window: 32K tokens

Deprecated by provider · still listed · price checked Nov 12, 2026 (deprecation notice ↗)

All rates, conditions and sources

Conditions of every rate below:

  • Listed for model "Voxtral Small 1.0" (provider "Mistral") in us-east-1 with service_tier "standard", as one USD price dimension from 0 to Inf.
  • AWS lists this price in USD per 1K tokens (unit "1K tokens"); it is multiplied by 1,000 exactly, by decimal arithmetic, to give USD per 1M tokens.
  • Only the standard on-demand tier in us-east-1 is published: flex, priority, batch and global cross-region tiers, other AWS Regions and provisioned throughput are excluded. The offer file lists no standard-tier cache price for this model; missing cache prices are unknown, not zero.

Input: $0.10 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-mistral.voxtral-small-24b-2507-mantle-input-tokens-standard.
  • Input tokens.

Output: $0.30 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-mistral.voxtral-small-24b-2507-mantle-output-tokens-standard.
  • Output tokens.

Price observed:

API availability: listed as available when checked (not a live service check)

Collection record

Primary source ↗ · provider documentation · Claims: bedrock-voxtral-small-24b-2507: specifications · Retrieved · Checked · Reviewed · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card ("Models at a glance") for Voxtral Small 24B 2507, Model ID mistral.voxtral-small-24b-2507 (the offer's price list usagetype stem), raw SHA-256 37101155fa55ba3ac483df94b1342ce876e2efd3a1aa971dc91271f6a0730d4e, read once on 2026-10-31 under the #837 AWS facts-plus-link basis: "Context window: 32K tokens".
  • Shorthand recorded as stated (#504); not converted to an exact token count. The value is AWS's statement for its Bedrock listing; another seller of the same model may state a different figure.
  • Specifications only: no pricing, availability, capability, lifecycle or release-date assertion.

Primary source ↗ · provider documentation · Claims: bedrock-voxtral-small-24b-2507: availability · Retrieved · Checked · Reviewed · Review due · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card for Voxtral Small 24B 2507, Model ID mistral.voxtral-small-24b-2507, raw SHA-256 37101155fa55ba3ac483df94b1342ce876e2efd3a1aa971dc91271f6a0730d4e, read once on 2026-10-31 under the #837 AWS facts-plus-link basis: "Model lifecycle: Legacy"; "Model EOL date: March 30, 2027"; "Model lifecycle policy: Model lifecycle (For Models Launched Prior to Sept 7 2026)".
  • Amazon Bedrock "Model lifecycle (Legacy)" page (https://docs.aws.amazon.com/bedrock/latest/userguide/model-lifecycle-legacy.html), read once at 2026-10-31T16:36:03Z, raw SHA-256 c33d9d25d7da7f47c098aa864830361de770224f38a790070bada021dac56850: its Legacy table lists Voxtral Small 24B 2507 (mistral.voxtral-small-24b-2507) with Legacy date September 30, 2026, EOL date March 30, 2027 and public extended access start date "To be announced". The page states: "New customers can't use Legacy models and existing customers may lose access to Legacy models after 15 days of inactivity."
  • That table row's Regions include us-east-1, the region this offer is published for.
  • Availability stays available: the us-east-1 price list still lists an on-demand price, and the page says "you can continue to use a Legacy model during this period" (the Legacy period before the EOL date), though not as a new customer. The EOL date is in the future, so no retirement is recorded; reviewAfter is the EOL date, a re-review date, not a retirement.
  • Reviewed by Claude (owner delegation 2026-09-27), #909.

Raw API evidence ↗ · provider API · Claims: bedrock-voxtral-small-24b-2507: pricing · bedrock-voxtral-small-24b-2507: availability · 122 more claims about 61 other catalog entries · Retrieved · Checked · Review due · Confidence: A

  • Raw SHA-256: f1ac6a175fb27b3ad80d2c5d8970be8c2e5f957906349e52d15ec11d43e97896; collector aws-bedrock-price-list-json-1.
  • AWS Price List Bulk API offer file for Amazon Bedrock (us-east-1): offer version 20261111013037, publicationDate 2026-11-11T01:30:37Z.
  • Reviewed on-demand token prices only: each curated model's exact usagetypes in us-east-1, with service_tier "standard" for standard-tier models and with no service_tier attribute for the reviewed untiered models, in US dollars as the offer file states them (hence confidence A). A hosted listing and pricing observation only: no checkpoint, developer, license, context-window, modality or release-date assertions, and the usagetype is not a documented Bedrock model ID.
  • No Anthropic Claude price is published from this offer file: prices are read only from curated usagetypes, and no curated entry is an Anthropic model. (When the collector was reviewed on 2026-10-29, the file listed only legacy Claude models, all without a service tier.)
  • Availability rests on AWS's own offer file listing an on-demand price for the model in us-east-1.

Writer Palmyra Vision 7B

Context window: 4K tokens

All rates, conditions and sources

Conditions of every rate below:

  • Listed for model "Writer Palmyra Vision 7B" (provider "Writer") in us-east-1 with service_tier "standard", as one USD price dimension from 0 to Inf.
  • AWS lists this price in USD per 1K tokens (unit "1K tokens"); it is multiplied by 1,000 exactly, by decimal arithmetic, to give USD per 1M tokens.
  • Only the standard on-demand tier in us-east-1 is published: flex, priority, batch and global cross-region tiers, other AWS Regions and provisioned throughput are excluded. The offer file lists no standard-tier cache price for this model; missing cache prices are unknown, not zero.

Input: $0.15 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-writer.palmyra-vision-7b-mantle-input-tokens-standard.
  • Input tokens.

Output: $0.60 per 1M tokens · Standard tier · Region: us-east-1

  • AWS Price List Bulk API offer file for Amazon Bedrock in US East (N. Virginia) (pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json), on-demand (OnDemand) term of usagetype USE1-writer.palmyra-vision-7b-mantle-output-tokens-standard.
  • Output tokens.

Price observed:

API availability: listed as available when checked (not a live service check)

Collection record

Primary source ↗ · provider documentation · Claims: bedrock-palmyra-vision-7b: specifications · Retrieved · Checked · Reviewed · Freshness: unknown · Confidence: A

  • Official Amazon Bedrock model card ("Models at a glance") for Palmyra Vision 7B, Model ID writer.palmyra-vision-7b (the offer's price list usagetype stem), raw SHA-256 80702502f3898a6358466490f525b2fd119f28810f8c5e26b848295cc1a995b4, read once on 2026-10-31 under the #837 AWS facts-plus-link basis: "Context window: 4K tokens".
  • Shorthand recorded as stated (#504); not converted to an exact token count. The value is AWS's statement for its Bedrock listing; another seller of the same model may state a different figure.
  • Specifications only: no pricing, availability, capability, lifecycle or release-date assertion.

Raw API evidence ↗ · provider API · Claims: bedrock-palmyra-vision-7b: pricing · bedrock-palmyra-vision-7b: availability · 122 more claims about 61 other catalog entries · Retrieved · Checked · Review due · Confidence: A

  • Raw SHA-256: f1ac6a175fb27b3ad80d2c5d8970be8c2e5f957906349e52d15ec11d43e97896; collector aws-bedrock-price-list-json-1.
  • AWS Price List Bulk API offer file for Amazon Bedrock (us-east-1): offer version 20261111013037, publicationDate 2026-11-11T01:30:37Z.
  • Reviewed on-demand token prices only: each curated model's exact usagetypes in us-east-1, with service_tier "standard" for standard-tier models and with no service_tier attribute for the reviewed untiered models, in US dollars as the offer file states them (hence confidence A). A hosted listing and pricing observation only: no checkpoint, developer, license, context-window, modality or release-date assertions, and the usagetype is not a documented Bedrock model ID.
  • No Anthropic Claude price is published from this offer file: prices are read only from curated usagetypes, and no curated entry is an Anthropic model. (When the collector was reviewed on 2026-10-29, the file listed only legacy Claude models, all without a service tier.)
  • Availability rests on AWS's own offer file listing an on-demand price for the model in us-east-1.
About capability scores (Epoch AI)

Capability scores: Epoch AI, 'Capabilities & benchmarking'. Published online at epoch.ai. Retrieved from 'https://epoch.ai/benchmarks' [online resource], epoch.ai/benchmarks. Accessed Nov 18, 2026. Licensed CC BY 4.0; values here are rounded to one decimal place and grouped into bands.

The Epoch Capabilities Index (ECI) combines many benchmarks on a scale where Claude 3.5 Sonnet is 130 and GPT-5 is 150. Bands, on the displayed one-decimal score: Frontier 160+, Strong 150–159.9, Capable 140–149.9, Entry below 140. Each score has a 90% range, listed with each model's rates and sources; a range that crosses a band boundary is flagged there. Models are not ranked: neighbouring scores often overlap.

A benchmark is not your workload. Epoch's scores mostly reflect runs at the highest reasoning effort, which use more output tokens than default settings. Scores are dated and can shift between Epoch releases. A score belongs to the developer's model; a host listing shows the same model's score, but hosts may serve different versions or quantizations.

How headline rates are chosen

Headline rates use the directory's shared rule: the rate a buyer meets by default, chosen by billing, tier and context dimensions and never by price. Context caps, the bracket starting at zero and short-context variants are labeled. Every price is a listed price, not a service guarantee; each offer's details say what its source states about availability. Batch, off-peak and other variants remain in the details; where no single rate is the default, the table shows a dash.

No currency conversion is performed. Missing cache prices are unknown, not zero. Equal units do not establish equivalent models or service conditions. See the methodology and source register.