Claude Haiku 4.5 API pricing on Anthropic

Primary source ↗ · vendor-docs · Claims: anthropic-claude-haiku-4-5: identity · anthropic-claude-haiku-4-5: pricing · anthropic-claude-haiku-4-5: availability · anthropic-claude-haiku-4-5: specifications · Retrieved 2026-08-17T00:00:00Z · Effective: Unknown · Checked: Unknown · Reviewed: Unknown · Review after: Unknown · Freshness: unknown · Confidence: A

• Cached input is the cache-hit/refresh rate; cache writes cost more.

Primary source ↗ · vendor-docs · Claims: anthropic-claude-haiku-4-5: weightAccess · Retrieved 2026-09-08T05:17:14.115377Z · Effective: Unknown · Checked: Unknown · Reviewed: Unknown · Review after: Unknown · Freshness: unknown · Confidence: Unknown

• The official Claude Haiku 4.5 overview identifies the model and lists hosted platforms; this review did not establish explicit weight-release or non-downloadability evidence.
• Review outcome: unknown. No explicit model-specific weight-availability statement was established in the reviewed source; this is not evidence of non-downloadability. No download URL or weight license is asserted.

Primary source ↗ · vendor-docs · Claims: anthropic-claude-fable-5: pricing · anthropic-claude-haiku-4-5: pricing · anthropic-claude-opus-5: pricing · anthropic-claude-sonnet-5: pricing · anthropic-claude-opus-5-5: pricing · anthropic-claude-fable-5-1: pricing · Retrieved 2026-09-26T05:34:42.597Z · Effective: Unknown · Checked: 2026-09-28T04:40:19.422Z · Reviewed: Unknown · Review after: Unknown · Freshness: unknown · Confidence: A

• Raw SHA-256: cd36fcc6d6d9e17ac2a8643a3f769e42ec6cf33a8becdbfd98bc3ba8832f1aee; collector anthropic-markdown-2.
• Pricing observation only. Display-name mapping is not official API identity evidence. No new access, availability, weight, retirement or model metadata assertion.
• This page provides detailed pricing information for Anthropic's models and features. All prices are in USD. For the most current pricing information, visit claude.com/pricing.
• The following table shows pricing for all Claude models: 1 Cache hits and refreshes on Claude Fable 5.1 and Claude Mythos 5.1 are priced at 0.025x the base input price. 2 Cache hits and refreshes on Claude Opus 5.5 are priced at 0.05x the base input price. All other models use the standard 0.1x multiplier. 3 The $2/$10 per million input/output token pricing for Claude Sonnet 5, announced at launch as introductory pricing through August 31, 2026, is now the standard price. The previously scheduled increase to $3/$15 per million input/output tokens on September 1, 2026 will not occur. MTok: Million tokens. $5 / MTok is $5 for every million tokens. 5m cache writes: Writing a prompt prefix to the 5-minute prompt cache. 1h cache writes: Writing a prompt prefix to the 1-hour prompt cache. Cache hits and refreshes: Reading a prompt prefix from the prompt cache, which also refreshes it. Limited access: Offered separately, by invitation only, as part of Project Glasswing. For access, contact your Anthropic, AWS, or Google Cloud account team. Retired: May still be available on other cloud platforms. See Model deprecations for more. Claude 4.7 and later models and Claude Mythos Preview use a newer tokenizer that contributes to their improved performance on a wide range of tasks. This tokenizer produces approximately 30% more tokens for the same text. The exact increase depends on the content and workload shape. Claude Sonnet 4.6 and earlier models use the previous tokenizer.
• Prompt caching reduces costs and latency by reusing previously processed portions of your prompt across API calls. Instead of reprocessing the same large system prompt, document, or conversation history on every request, the API reads from cache at a fraction of the standard input price. There are two ways to enable prompt caching: Automatic caching: Add a single cache_control field at the top level of your request. The system automatically manages cache breakpoints as conversations grow. This is the recommended starting point for most use cases. Explicit cache breakpoints: Place cache_control directly on individual content blocks for fine-grained control over exactly what gets cached. Prompt caching uses the following pricing multipliers relative to base input token rates: Cache operation Multiplier Duration 5-minute cache write 1.25x base input price Cache valid for 5 minutes 1-hour cache write 2x base input price Cache valid for 1 hour Cache read (hit) 0.1x base input price (0.025x on Claude Fable 5.1 and Claude Mythos 5.1; 0.05x on Claude Opus 5.5) Same duration as the preceding write Cache write tokens are charged when content is first stored. Cache read tokens are charged when a subsequent request retrieves the cached content. A cache hit costs 10% of the standard input price, which means caching pays off after one cache read for the 5-minute duration (1.25x write), or after two cache reads for the 1-hour duration (2x write). On Claude Fable 5.1 and Claude Mythos 5.1, a cache hit costs 2.5% of the standard input price ($0.25 USD per million tokens). On Claude Opus 5.5, a cache hit costs 5% of the standard input price ($0.20 USD per million tokens). These multipliers stack with other pricing modifiers, including the Batch API discount and data residency. For implementation details, supported models, and code examples, see Prompt caching.
• For Claude 4.6 and later models, specifying US-only inference through the inference_geo parameter incurs a 1.1x multiplier on all token pricing categories, including input tokens, output tokens, cache writes, and cache reads. Global routing (the default) uses standard pricing. This applies to the Claude API (first-party) and Claude Platform on AWS. On Claude in Microsoft Foundry, the same 1.1x multiplier applies to deployments that use the US Data Zone Standard deployment type (see Inference geography). Partner-operated platforms (Bedrock and Google Cloud) have independent regional pricing. See Bedrock and Google Cloud for details. Earlier models do not support the inference_geo parameter and always use standard pricing; requests that include the parameter on these models return a 400 error. For more information, see Data residency.
• Claude 4.6 and later models and Claude Mythos Preview include the full 1M token context window at standard pricing. (A 900k-token request is billed at the same per-token rate as a 9k-token request.) Prompt caching and batch processing discounts apply at standard rates across the full context window.

Primary source ↗ · vendor-docs · Claims: anthropic-claude-fable-5: pricing · anthropic-claude-haiku-4-5: pricing · anthropic-claude-opus-5: pricing · anthropic-claude-sonnet-5: pricing · anthropic-claude-opus-5-5: pricing · anthropic-claude-fable-5-1: pricing · Retrieved 2026-09-29T04:37:41.581Z · Effective: Unknown · Checked: 2026-10-06T04:38:06.956Z · Reviewed: Unknown · Review after: Unknown · Freshness: unknown · Confidence: A

• Raw SHA-256: 245b5d0472659075015551126902de479990475fa61deb58c5fc20ef5e51a909; collector anthropic-markdown-2.
• Pricing observation only. Display-name mapping is not official API identity evidence. No new access, availability, weight, retirement or model metadata assertion.
• This page provides detailed pricing information for Anthropic's models and features. All prices are in USD. For the most current pricing information, visit claude.com/pricing.
• The following table shows pricing for all Claude models: 1 Cache hits and refreshes on Claude Fable 5.1 and Claude Mythos 5.1 are priced at 0.025x the base input price. 2 Cache hits and refreshes on Claude Opus 5.5 are priced at 0.05x the base input price. All other models use the standard 0.1x multiplier. 3 The $2/$10 per million input/output token pricing for Claude Sonnet 5, announced at launch as introductory pricing through August 31, 2026, is now the standard price. The previously scheduled increase to $3/$15 per million input/output tokens on September 1, 2026 will not occur. MTok: Million tokens. $5 / MTok is $5 for every million tokens. 5m cache writes: Writing a prompt prefix to the 5-minute prompt cache. 1h cache writes: Writing a prompt prefix to the 1-hour prompt cache. Cache hits and refreshes: Reading a prompt prefix from the prompt cache, which also refreshes it. Limited access: Offered separately, by invitation only, as part of Project Glasswing. For access, contact your Anthropic, AWS, or Google Cloud account team. Retired: May still be available on other cloud platforms. See Model deprecations for more. Claude 4.7 and later models and Claude Mythos Preview use a newer tokenizer that contributes to their improved performance on a wide range of tasks. This tokenizer produces approximately 30% more tokens for the same text. The exact increase depends on the content and workload shape. Claude Sonnet 4.6 and earlier models use the previous tokenizer.
• Prompt caching reduces costs and latency by reusing previously processed portions of your prompt across API calls. Instead of reprocessing the same large system prompt, document, or conversation history on every request, the API reads from cache at a fraction of the standard input price. There are two ways to enable prompt caching: Automatic caching: Add a single cache_control field at the top level of your request. The system automatically manages cache breakpoints as conversations grow. This is the recommended starting point for most use cases. Explicit cache breakpoints: Place cache_control directly on individual content blocks for fine-grained control over exactly what gets cached. Prompt caching uses the following pricing multipliers relative to base input token rates: Cache operation Multiplier Duration 5-minute cache write 1.25x base input price Cache valid for 5 minutes 1-hour cache write 2x base input price Cache valid for 1 hour Cache read (hit) 0.1x base input price (0.025x on Claude Fable 5.1 and Claude Mythos 5.1; 0.05x on Claude Opus 5.5) Same duration as the preceding write Cache write tokens are charged when content is first stored. Cache read tokens are charged when a subsequent request retrieves the cached content. A cache hit costs 10% of the standard input price, which means caching pays off after one cache read for the 5-minute duration (1.25x write), or after two cache reads for the 1-hour duration (2x write). On Claude Fable 5.1 and Claude Mythos 5.1, a cache hit costs 2.5% of the standard input price ($0.25 USD per million tokens). On Claude Opus 5.5, a cache hit costs 5% of the standard input price ($0.20 USD per million tokens). These multipliers stack with other pricing modifiers, including the Batch API discount and data residency. For implementation details, supported models, and code examples, see Prompt caching.
• For Claude 4.6 and later models, specifying US-only inference through the inference_geo parameter incurs a 1.1x multiplier on all token pricing categories, including input tokens, output tokens, cache writes, and cache reads. Global routing (the default) uses standard pricing. This applies to the Claude API (first-party) and Claude Platform on AWS. On Claude in Microsoft Foundry, the same 1.1x multiplier applies to deployments that use the US Data Zone Standard deployment type (see Inference geography). Partner-operated platforms (Bedrock and Google Cloud) have independent regional pricing. See Bedrock and Google Cloud for details. Earlier models do not support the inference_geo parameter and always use standard pricing; requests that include the parameter on these models return a 400 error. For more information, see Data residency.
• Claude 4.6 and later models and Claude Mythos Preview include the full 1M token context window at standard pricing. (A 900k-token request is billed at the same per-token rate as a 9k-token request.) Prompt caching and batch processing discounts apply at standard rates across the full context window.

Primary source ↗ · vendor-docs · Claims: anthropic-claude-fable-5: pricing · anthropic-claude-haiku-4-5: pricing · anthropic-claude-opus-5: pricing · anthropic-claude-sonnet-5: pricing · anthropic-claude-opus-5-5: pricing · anthropic-claude-fable-5-1: pricing · anthropic-claude-sonnet-5-5: pricing · anthropic-claude-haiku-5-5: pricing · Retrieved 2026-10-11T04:39:01.978Z · Effective: Unknown · Checked: 2026-10-11T04:39:01.978Z · Reviewed: Unknown · Review after: Unknown · Freshness: unknown · Confidence: A

• Raw SHA-256: 67904e280b595d31f373c857844ed380072ae79d280d8f385640b5992d337ca9; collector anthropic-markdown-2.
• Pricing observation only. Display-name mapping is not official API identity evidence; official IDs carry their own model-page identity records (#449). No new access, availability, weight, retirement or model metadata assertion.
• This page provides detailed pricing information for Anthropic's models and features. All prices are in USD. For the most current pricing information, visit claude.com/pricing.
• The following table shows pricing for all Claude models: 1 Cache hits and refreshes on Claude Fable 5.1 and Claude Mythos 5.1 are priced at 0.025x the base input price. 2 Cache hits and refreshes on Claude Opus 5.5 and Claude Sonnet 5.5 are priced at 0.05x the base input price. All other models use the standard 0.1x multiplier. 3 The $2/$10 per million input/output token pricing for Claude Sonnet 5, announced at launch as introductory pricing through August 31, 2026, is now the standard price. The previously scheduled increase to $3/$15 per million input/output tokens on September 1, 2026 will not occur. MTok: Million tokens. $5 / MTok is $5 for every million tokens. 5m cache writes: Writing a prompt prefix to the 5-minute prompt cache. 1h cache writes: Writing a prompt prefix to the 1-hour prompt cache. Cache hits and refreshes: Reading a prompt prefix from the prompt cache, which also refreshes it. Limited access: Available only to organizations verified through Anthropic’s verification programs, such as the Cyber Verification Program. To request access, apply to the program that covers your use case. Retired: May still be available on other cloud platforms. See Model deprecations for more. Claude 4.7 and later models and Claude Mythos Preview use a newer tokenizer that contributes to their improved performance on a wide range of tasks. This tokenizer produces approximately 30% more tokens for the same text. The exact increase depends on the content and workload shape. Claude Sonnet 4.6 and earlier models use the previous tokenizer.
• Prompt caching reduces costs and latency by reusing previously processed portions of your prompt across API calls. Instead of reprocessing the same large system prompt, document, or conversation history on every request, the API reads from cache at a fraction of the standard input price. There are two ways to enable prompt caching: Automatic caching: Add a single cache_control field at the top level of your request. The system automatically manages cache breakpoints as conversations grow. This is the recommended starting point for most use cases. Explicit cache breakpoints: Place cache_control directly on individual content blocks for fine-grained control over exactly what gets cached. Prompt caching uses the following pricing multipliers relative to base input token rates: Cache operation Multiplier Duration 5-minute cache write 1.25x base input price Cache valid for 5 minutes 1-hour cache write 2x base input price Cache valid for 1 hour Cache read (hit) 0.1x base input price (0.025x on Claude Fable 5.1 and Claude Mythos 5.1; 0.05x on Claude Opus 5.5 and Claude Sonnet 5.5) Same duration as the preceding write Cache write tokens are charged when content is first stored. Cache read tokens are charged when a subsequent request retrieves the cached content. A cache hit costs 10% of the standard input price, which means caching pays off after one cache read for the 5-minute duration (1.25x write), or after two cache reads for the 1-hour duration (2x write). On Claude Fable 5.1 and Claude Mythos 5.1, a cache hit costs 2.5% of the standard input price ($0.25 USD per million tokens). On Claude Opus 5.5 and Claude Sonnet 5.5, a cache hit costs 5% of the standard input price ($0.20 USD per million tokens on Claude Opus 5.5, $0.10 USD on Claude Sonnet 5.5). These multipliers stack with other pricing modifiers, including the Batch API discount and data residency. For implementation details, supported models, and code examples, see Prompt caching.
• For Claude 4.6 and later models, specifying US-only inference through the inference_geo parameter incurs a 1.1x multiplier on all token pricing categories, including input tokens, output tokens, cache writes, and cache reads. On Claude Haiku 5.5, the multiplier also applies to the higher prices for prompts over 100,000 tokens (see Long context pricing). Global routing (the default) uses standard pricing. This applies to the Claude API (first-party) and Claude Platform on AWS. On Claude in Microsoft Foundry, the same 1.1x multiplier applies to deployments that use the US Data Zone Standard deployment type (see Inference geography). Partner-operated platforms (Bedrock and Google Cloud) have independent regional pricing. See Bedrock and Google Cloud for details. Earlier models do not support the inference_geo parameter and always use standard pricing; requests that include the parameter on these models return a 400 error. For more information, see Data residency.
• Claude 4.6 and later models (except Claude Haiku 5.5) and Claude Mythos Preview include the full 1M token context window at standard pricing. (A 900k-token request is billed at the same per-token rate as a 9k-token request.) Prompt caching and batch processing discounts apply at standard rates across the full context window. Claude Haiku 5.5 is priced by prompt length: a request whose prompt is over 100,000 tokens pays higher prices. A request's prompt length counts all of its input tokens, including cache reads and cache writes. Each request is priced on its own: a request over the threshold pays the higher prices even when part of its prompt is a cache hit, and earlier requests keep the prices they were billed at. Model pricing and Batch processing list both sets of prices.

Primary source ↗ · vendor-docs · Claims: anthropic-claude-haiku-4-5: identity · Retrieved 2026-10-10T22:54:38Z · Effective: Unknown · Checked: 2026-10-10T22:54:38Z · Reviewed: Unknown · Review after: Unknown · Freshness: unknown · Confidence: A

• Official model page, Markdown rendition SHA-256 bec0ed34119af773736e4ee5897e25cf43b1261a5e63b77b0639fb36814f32e7. "Model IDs" table row: | Claude API | `claude-haiku-4-5-20251001` |
• The page also lists "| Claude API alias | `claude-haiku-4-5` |", a convenience pointer that resolves to this dated snapshot.
• Copied on 2026-10-10 for #449 (table-row whitespace collapsed) and checked in code review; no separate review timestamp is recorded. Identity only: no pricing, availability, context or capability assertion.

Primary source ↗ · vendor-docs · Claims: anthropic-claude-fable-5: pricing · anthropic-claude-haiku-4-5: pricing · anthropic-claude-opus-5: pricing · anthropic-claude-sonnet-5: pricing · anthropic-claude-opus-5-5: pricing · anthropic-claude-fable-5-1: pricing · anthropic-claude-sonnet-5-5: pricing · anthropic-claude-haiku-5-5: pricing · Retrieved 2026-10-14T04:36:53.232Z · Effective: Unknown · Checked: 2026-10-14T04:36:53.232Z · Reviewed: Unknown · Review after: Unknown · Freshness: unknown · Confidence: A

• Raw SHA-256: 9c7be951e649665896711d46691b191ef0d4e0ba29b937091053c808393abd63; collector anthropic-markdown-2.
• Pricing observation only. Display-name mapping is not official API identity evidence; official IDs carry their own model-page identity records (#449). No new access, availability, weight, retirement or model metadata assertion.
• This page provides detailed pricing information for Anthropic's models and features. All prices are in USD. For the most current pricing information, visit claude.com/pricing.
• The following table shows pricing for all Claude models: 1 Cache hits and refreshes on Claude Fable 5.1 and Claude Mythos 5.1 are priced at 0.025x the base input price. 2 Cache hits and refreshes on Claude Opus 5.5 and Claude Sonnet 5.5 are priced at 0.05x the base input price. All other models use the standard 0.1x multiplier. 3 The $2/$10 per million input/output token pricing for Claude Sonnet 5, announced at launch as introductory pricing through August 31, 2026, is now the standard price. The previously scheduled increase to $3/$15 per million input/output tokens on September 1, 2026 will not occur. MTok: Million tokens. $5 / MTok is $5 for every million tokens. 5m cache writes: Writing a prompt prefix to the 5-minute prompt cache. 1h cache writes: Writing a prompt prefix to the 1-hour prompt cache. Cache hits and refreshes: Reading a prompt prefix from the prompt cache, which also refreshes it. Limited access: Available only to organizations verified through Anthropic’s verification programs, such as the Cyber Verification Program. To request access, apply to the program that covers your use case. Retired: May still be available on other cloud platforms. See Model deprecations for more. Claude 4.7 and later models and Claude Mythos Preview use a newer tokenizer that contributes to their improved performance on a wide range of tasks. This tokenizer produces approximately 30% more tokens for the same text. The exact increase depends on the content and workload shape. Claude Sonnet 4.6 and earlier models use the previous tokenizer.
• Prompt caching reduces costs and latency by reusing previously processed portions of your prompt across API calls. Instead of reprocessing the same large system prompt, document, or conversation history on every request, the API reads from cache at a fraction of the standard input price. There are two ways to enable prompt caching: Automatic caching: Add a single cache_control field at the top level of your request. The system automatically manages cache breakpoints as conversations grow. This is the recommended starting point for most use cases. Explicit cache breakpoints: Place cache_control directly on individual content blocks for fine-grained control over exactly what gets cached. Prompt caching uses the following pricing multipliers relative to base input token rates: Cache operation Multiplier Duration 5-minute cache write 1.25x base input price Cache valid for 5 minutes 1-hour cache write 2x base input price Cache valid for 1 hour Cache read (hit) 0.1x base input price (0.025x on Claude Fable 5.1 and Claude Mythos 5.1; 0.05x on Claude Opus 5.5 and Claude Sonnet 5.5) Same duration as the preceding write Cache write tokens are charged when content is first stored. Cache read tokens are charged when a subsequent request retrieves the cached content. A cache hit costs 10% of the standard input price, which means caching pays off after one cache read for the 5-minute duration (1.25x write), or after two cache reads for the 1-hour duration (2x write). On Claude Fable 5.1 and Claude Mythos 5.1, a cache hit costs 2.5% of the standard input price ($0.25 USD per million tokens). On Claude Opus 5.5 and Claude Sonnet 5.5, a cache hit costs 5% of the standard input price ($0.20 USD per million tokens on Claude Opus 5.5, $0.10 USD on Claude Sonnet 5.5). These multipliers stack with other pricing modifiers, including the Batch API discount and data residency. For implementation details, supported models, and code examples, see Prompt caching.
• For Claude 4.6 and later models, specifying US-only inference through the inference_geo parameter incurs a 1.1x multiplier on all token pricing categories, including input tokens, output tokens, cache writes, and cache reads. On Claude Haiku 5.5, the multiplier also applies to the higher prices for prompts over 100,000 tokens (see Long context pricing). Global routing (the default) uses standard pricing. This applies to the Claude API (first-party) and Claude Platform on AWS. On Claude in Microsoft Foundry, the same 1.1x multiplier applies to deployments that use the US Data Zone Standard deployment type (see Inference geography). Partner-operated platforms (Bedrock and Google Cloud) have independent regional pricing. See Bedrock and Google Cloud for details. Earlier models do not support the inference_geo parameter and always use standard pricing; requests that include the parameter on these models return a 400 error. For more information, see Data residency.
• Claude 4.6 and later models (except Claude Haiku 5.5) and Claude Mythos Preview include the full 1M token context window at standard pricing. (A 900k-token request is billed at the same per-token rate as a 9k-token request.) Prompt caching and batch processing discounts apply at standard rates across the full context window. Claude Haiku 5.5 is priced by prompt length: a request whose prompt is over 100,000 tokens pays higher prices. A request's prompt length counts all of its input tokens, including cache reads and cache writes. Each request is priced on its own: a request over the threshold pays the higher prices even when part of its prompt is a cache hit, and earlier requests keep the prices they were billed at. Model pricing and Batch processing list both sets of prices.

Primary source ↗ · vendor-docs · Claims: anthropic-claude-fable-5: pricing · anthropic-claude-haiku-4-5: pricing · anthropic-claude-opus-5: pricing · anthropic-claude-sonnet-5: pricing · anthropic-claude-opus-5-5: pricing · anthropic-claude-fable-5-1: pricing · anthropic-claude-sonnet-5-5: pricing · anthropic-claude-haiku-5-5: pricing · Retrieved 2026-10-15T04:34:55.067Z · Effective: Unknown · Checked: 2026-10-15T04:34:55.067Z · Reviewed: Unknown · Review after: 2026-10-25T04:34:55.067Z · Freshness: stale · Confidence: A

• Raw SHA-256: a62e85bd0420e44f326f7b6c6c6d9fdb879dbf0e513b93b9f671b72caae5c83b; collector anthropic-markdown-2.
• Pricing observation only. Display-name mapping is not official API identity evidence; official IDs carry their own model-page identity records (#449). No new access, availability, weight, retirement or model metadata assertion.
• This page provides detailed pricing information for Anthropic's models and features. All prices are in USD. For the most current pricing information, visit claude.com/pricing.
• The following table shows pricing for all Claude models: 1 Cache hits and refreshes on Claude Fable 5.1 and Claude Mythos 5.1 are priced at 0.025x the base input price. 2 Cache hits and refreshes on Claude Opus 5.5 and Claude Sonnet 5.5 are priced at 0.05x the base input price. All other models use the standard 0.1x multiplier. 3 The $2/$10 per million input/output token pricing for Claude Sonnet 5, announced at launch as introductory pricing through August 31, 2026, is now the standard price. The previously scheduled increase to $3/$15 per million input/output tokens on September 1, 2026 will not occur. MTok: Million tokens. $5 / MTok is $5 for every million tokens. 5m cache writes: Writing a prompt prefix to the 5-minute prompt cache. 1h cache writes: Writing a prompt prefix to the 1-hour prompt cache. Cache hits and refreshes: Reading a prompt prefix from the prompt cache, which also refreshes it. Limited access: Available only to organizations verified through Anthropic’s verification programs, such as the Cyber Verification Program. To request access, apply to the program that covers your use case. Retired: May still be available on other cloud platforms. See Model deprecations for more. Claude 4.7 and later models and Claude Mythos Preview use a newer tokenizer that contributes to their improved performance on a wide range of tasks. This tokenizer produces approximately 30% more tokens for the same text. The exact increase depends on the content and workload shape. Claude Sonnet 4.6 and earlier models use the previous tokenizer.
• Prompt caching reduces costs and latency by reusing previously processed portions of your prompt across API calls. Instead of reprocessing the same large system prompt, document, or conversation history on every request, the API reads from cache at a fraction of the standard input price. There are two ways to enable prompt caching: Automatic caching: Add a single cache_control field at the top level of your request. The system automatically manages cache breakpoints as conversations grow. This is the recommended starting point for most use cases. Explicit cache breakpoints: Place cache_control directly on individual content blocks for fine-grained control over exactly what gets cached. Prompt caching uses the following pricing multipliers relative to base input token rates: Cache operation Multiplier Duration 5-minute cache write 1.25x base input price Cache valid for 5 minutes 1-hour cache write 2x base input price Cache valid for 1 hour Cache read (hit) 0.1x base input price (0.025x on Claude Fable 5.1 and Claude Mythos 5.1; 0.05x on Claude Opus 5.5 and Claude Sonnet 5.5) Same duration as the preceding write Cache write tokens are charged when content is first stored. Cache read tokens are charged when a subsequent request retrieves the cached content. A cache hit costs 10% of the standard input price, which means caching pays off after one cache read for the 5-minute duration (1.25x write), or after two cache reads for the 1-hour duration (2x write). On Claude Fable 5.1 and Claude Mythos 5.1, a cache hit costs 2.5% of the standard input price ($0.25 USD per million tokens). On Claude Opus 5.5 and Claude Sonnet 5.5, a cache hit costs 5% of the standard input price ($0.20 USD per million tokens on Claude Opus 5.5, $0.10 USD on Claude Sonnet 5.5). These multipliers stack with other pricing modifiers, including the Batch API discount and data residency. For implementation details, supported models, and code examples, see Prompt caching.
• For Claude 4.6 and later models, specifying US-only inference through the inference_geo parameter incurs a 1.1x multiplier on all token pricing categories, including input tokens, output tokens, cache writes, and cache reads. On Claude Haiku 5.5, the multiplier also applies to the higher prices for prompts over 100,000 tokens (see Long context pricing). Global routing (the default) uses standard pricing. This applies to the Claude API (first-party) and Claude Platform on AWS. On Claude in Microsoft Foundry, the same 1.1x multiplier applies to deployments that use the US Data Zone Standard deployment type (see Inference geography). Partner-operated platforms (Bedrock and Google Cloud) have independent regional pricing. See Bedrock and Google Cloud for details. Earlier models do not support the inference_geo parameter and always use standard pricing; requests that include the parameter on these models return a 400 error. For more information, see Data residency.
• Claude 4.6 and later models (except Claude Haiku 5.5) and Claude Mythos Preview include the full 1M token context window at standard pricing. (A 900k-token request is billed at the same per-token rate as a 9k-token request.) Prompt caching and batch processing discounts apply at standard rates across the full context window. Claude Haiku 5.5 is priced by prompt length: a request whose prompt is over 100,000 tokens pays higher prices. A request's prompt length counts all of its input tokens, including cache reads and cache writes. Each request is priced on its own: a request over the threshold pays the higher prices even when part of its prompt is a cache hit, and earlier requests keep the prices they were billed at. Model pricing and Batch processing list both sets of prices.

Primary source ↗ · vendor-docs · Claims: anthropic-claude-fable-5: pricing · anthropic-claude-haiku-4-5: pricing · anthropic-claude-opus-5: pricing · anthropic-claude-sonnet-5: pricing · anthropic-claude-opus-5-5: pricing · anthropic-claude-fable-5-1: pricing · anthropic-claude-sonnet-5-5: pricing · anthropic-claude-haiku-5-5: pricing · Retrieved 2026-10-16T04:35:04.708Z · Effective: Unknown · Checked: 2026-10-16T04:35:04.708Z · Reviewed: Unknown · Review after: 2026-10-26T04:35:04.708Z · Freshness: stale · Confidence: A

• Raw SHA-256: d0dc6009ff1bae0023eec432d7c21d4a3b71a71e542ac350dbfd13dc85a8b862; collector anthropic-markdown-2.
• Pricing observation only. Display-name mapping is not official API identity evidence; official IDs carry their own model-page identity records (#449). No new access, availability, weight, retirement or model metadata assertion.
• This page provides detailed pricing information for Anthropic's models and features. All prices are in USD. For the most current pricing information, visit claude.com/pricing.
• The following table shows pricing for all Claude models: 1 Cache hits and refreshes on Claude Fable 5.1 and Claude Mythos 5.1 are priced at 0.025x the base input price. 2 Cache hits and refreshes on Claude Opus 5.5 and Claude Sonnet 5.5 are priced at 0.05x the base input price. All other models use the standard 0.1x multiplier. 3 The $2/$10 per million input/output token pricing for Claude Sonnet 5, announced at launch as introductory pricing through August 31, 2026, is now the standard price. The previously scheduled increase to $3/$15 per million input/output tokens on September 1, 2026 will not occur. MTok: Million tokens. $5 / MTok is $5 for every million tokens. 5m cache writes: Writing a prompt prefix to the 5-minute prompt cache. 1h cache writes: Writing a prompt prefix to the 1-hour prompt cache. Cache hits and refreshes: Reading a prompt prefix from the prompt cache, which also refreshes it. Limited access: Available only to organizations verified through Anthropic’s verification programs, such as the Cyber Verification Program. To request access, apply to the program that covers your use case. Retired: May still be available on other cloud platforms. See Model deprecations for more. Claude 4.7 and later models and Claude Mythos Preview use a newer tokenizer that contributes to their improved performance on a wide range of tasks. This tokenizer produces approximately 30% more tokens for the same text. The exact increase depends on the content and workload shape. Claude Sonnet 4.6 and earlier models use the previous tokenizer.
• Prompt caching reduces costs and latency by reusing previously processed portions of your prompt across API calls. Instead of reprocessing the same large system prompt, document, or conversation history on every request, the API reads from cache at a fraction of the standard input price. There are two ways to enable prompt caching: Automatic caching: Add a single cache_control field at the top level of your request. The system automatically manages cache breakpoints as conversations grow. This is the recommended starting point for most use cases. Explicit cache breakpoints: Place cache_control directly on individual content blocks for fine-grained control over exactly what gets cached. Prompt caching uses the following pricing multipliers relative to base input token rates: Cache operation Multiplier Duration 5-minute cache write 1.25x base input price Cache valid for 5 minutes 1-hour cache write 2x base input price Cache valid for 1 hour Cache read (hit) 0.1x base input price (0.025x on Claude Fable 5.1 and Claude Mythos 5.1; 0.05x on Claude Opus 5.5 and Claude Sonnet 5.5) Same duration as the preceding write Cache write tokens are charged when content is first stored. Cache read tokens are charged when a subsequent request retrieves the cached content. A cache hit costs 10% of the standard input price, which means caching pays off after one cache read for the 5-minute duration (1.25x write), or after two cache reads for the 1-hour duration (2x write). On Claude Fable 5.1 and Claude Mythos 5.1, a cache hit costs 2.5% of the standard input price ($0.25 USD per million tokens). On Claude Opus 5.5 and Claude Sonnet 5.5, a cache hit costs 5% of the standard input price ($0.20 USD per million tokens on Claude Opus 5.5, $0.10 USD on Claude Sonnet 5.5). These multipliers stack with other pricing modifiers, including the Batch API discount and data residency. For implementation details, supported models, and code examples, see Prompt caching.
• For Claude 4.6 and later models, specifying US-only inference through the inference_geo parameter incurs a 1.1x multiplier on all token pricing categories, including input tokens, output tokens, cache writes, and cache reads. On Claude Haiku 5.5, the multiplier also applies to the higher prices for prompts over 100,000 tokens (see Long context pricing). Global routing (the default) uses standard pricing. This applies to the Claude API (first-party) and Claude Platform on AWS. On Claude in Microsoft Foundry, the same 1.1x multiplier applies to deployments that use the US Data Zone Standard deployment type (see Inference geography). Partner-operated platforms (Bedrock and Google Cloud) have independent regional pricing. See Bedrock and Google Cloud for details. Earlier models do not support the inference_geo parameter and always use standard pricing; requests that include the parameter on these models return a 400 error. For more information, see Data residency.
• Claude 4.6 and later models (except Claude Haiku 5.5) and Claude Mythos Preview include the full 1M token context window at standard pricing. (A 900k-token request is billed at the same per-token rate as a 9k-token request.) Prompt caching and batch processing discounts apply at standard rates across the full context window. Claude Haiku 5.5 is priced by prompt length: a request whose prompt is over 100,000 tokens pays higher prices. A request's prompt length counts all of its input tokens, including cache reads and cache writes. Each request is priced on its own: a request over the threshold pays the higher prices even when part of its prompt is a cache hit, and earlier requests keep the prices they were billed at. Model pricing and Batch processing list both sets of prices.

Primary source ↗ · vendor-docs · Claims: anthropic-claude-haiku-4-5: specifications · Retrieved 2026-10-14T01:44:45Z · Effective: Unknown · Checked: 2026-10-14T01:44:45Z · Reviewed: 2026-10-17T22:41:12Z · Review after: Unknown · Freshness: unknown · Confidence: A

• Official Anthropic model overview page for claude-haiku-4-5-20251001, raw SHA-256 bec0ed34119af773736e4ee5897e25cf43b1261a5e63b77b0639fb36814f32e7: "Context window: 200K tokens".
• Visible shorthand recorded as stated (#504); not converted to an exact token count.
• Specifications only: no pricing, availability, capability or release-date assertion.

Primary source ↗ · vendor-docs · Claims: anthropic-claude-fable-5: pricing · anthropic-claude-haiku-4-5: pricing · anthropic-claude-opus-5: pricing · anthropic-claude-sonnet-5: pricing · anthropic-claude-opus-5-5: pricing · anthropic-claude-fable-5-1: pricing · anthropic-claude-sonnet-5-5: pricing · anthropic-claude-haiku-5-5: pricing · Retrieved 2026-10-26T01:00:04.718Z · Effective: Unknown · Checked: 2026-10-26T01:00:04.718Z · Reviewed: Unknown · Review after: 2026-11-05T01:00:04.718Z · Freshness: stale · Confidence: A

• Raw SHA-256: f71573e3b8280316d567a310cdc9a469c6c56a3a0c13d9156446e4af340be39e; collector anthropic-markdown-2.
• Pricing observation only. Display-name mapping is not official API identity evidence; official IDs carry their own model-page identity records (#449). No new access, availability, weight, retirement or model metadata assertion.
• This page provides detailed pricing information for Anthropic's models and features. All prices are in USD. For the most current pricing information, visit claude.com/pricing.
• The following table shows pricing for all Claude models: 1 Cache hits and refreshes on Claude Fable 5.1 and Claude Mythos 5.1 are priced at 0.025x the base input price. 2 Cache hits and refreshes on Claude Opus 5.5 and Claude Sonnet 5.5 are priced at 0.05x the base input price. All other models use the standard 0.1x multiplier. 3 The $2/$10 per million input/output token pricing for Claude Sonnet 5, announced at launch as introductory pricing through August 31, 2026, is now the standard price. The previously scheduled increase to $3/$15 per million input/output tokens on September 1, 2026 will not occur. MTok: Million tokens. $5 / MTok is $5 for every million tokens. 5m cache writes: Writing a prompt prefix to the 5-minute prompt cache. 1h cache writes: Writing a prompt prefix to the 1-hour prompt cache. Cache hits and refreshes: Reading a prompt prefix from the prompt cache, which also refreshes it. Limited access: Available only to organizations verified through Anthropic’s verification programs, such as the Cyber Verification Program. To request access, apply to the program that covers your use case. Retired: May still be available on other cloud platforms. See Model deprecations for more. Claude 4.7 and later models and Claude Mythos Preview use a newer tokenizer that contributes to their improved performance on a wide range of tasks. This tokenizer produces approximately 30% more tokens for the same text. The exact increase depends on the content and workload shape. Claude Sonnet 4.6 and earlier models use the previous tokenizer.
• Prompt caching reduces costs and latency by reusing previously processed portions of your prompt across API calls. Instead of reprocessing the same large system prompt, document, or conversation history on every request, the API reads from cache at a fraction of the standard input price. There are two ways to enable prompt caching: Automatic caching: Add a single cache_control field at the top level of your request. The system automatically manages cache breakpoints as conversations grow. This is the recommended starting point for most use cases. Explicit cache breakpoints: Place cache_control directly on individual content blocks for fine-grained control over exactly what gets cached. Prompt caching uses the following pricing multipliers relative to base input token rates: Cache operation Multiplier Duration 5-minute cache write 1.25x base input price Cache valid for 5 minutes 1-hour cache write 2x base input price Cache valid for 1 hour Cache read (hit) 0.1x base input price (0.025x on Claude Fable 5.1 and Claude Mythos 5.1; 0.05x on Claude Opus 5.5 and Claude Sonnet 5.5) Same duration as the preceding write Cache write tokens are charged when content is first stored. Cache read tokens are charged when a subsequent request retrieves the cached content. A cache hit costs 10% of the standard input price, which means caching pays off after one cache read for the 5-minute duration (1.25x write), or after two cache reads for the 1-hour duration (2x write). On Claude Fable 5.1 and Claude Mythos 5.1, a cache hit costs 2.5% of the standard input price ($0.25 USD per million tokens). On Claude Opus 5.5 and Claude Sonnet 5.5, a cache hit costs 5% of the standard input price ($0.20 USD per million tokens on Claude Opus 5.5, $0.10 USD on Claude Sonnet 5.5). These multipliers stack with other pricing modifiers, including the Batch API discount and data residency. For implementation details, supported models, and code examples, see Prompt caching.
• For Claude 4.6 and later models, specifying US-only inference through the inference_geo parameter incurs a 1.1x multiplier on all token pricing categories, including input tokens, output tokens, cache writes, and cache reads. On Claude Haiku 5.5, the multiplier also applies to the higher prices for prompts over 100,000 tokens (see Long context pricing). Global routing (the default) uses standard pricing. This applies to the Claude API (first-party), Claude Platform on AWS, and Claude Models on Google Cloud (Anthropic managed). On Claude in Microsoft Foundry, the same 1.1x multiplier applies to deployments that use the US Data Zone Standard deployment type (see Inference geography). Partner-operated platforms, Bedrock and Claude Models on Google Cloud (Google managed), have independent regional pricing. See Bedrock pricing and Google Cloud pricing for details. Earlier models do not support the inference_geo parameter and always use standard pricing; requests that include the parameter on these models return a 400 error. For more information, see Data residency.
• Claude 4.6 and later models (except Claude Haiku 5.5) and Claude Mythos Preview include the full 1M token context window at standard pricing. (A 900k-token request is billed at the same per-token rate as a 9k-token request.) Prompt caching and batch processing discounts apply at standard rates across the full context window. Claude Haiku 5.5 is priced by prompt length: a request whose prompt is over 100,000 tokens pays higher prices. A request's prompt length counts all of its input tokens, including cache reads and cache writes. Each request is priced on its own: a request over the threshold pays the higher prices even when part of its prompt is a cache hit, and earlier requests keep the prices they were billed at. The model pricing and batch processing pricing sections list both sets of prices.

Primary source ↗ · vendor-docs · Claims: anthropic-claude-fable-5: pricing · anthropic-claude-haiku-4-5: pricing · anthropic-claude-opus-5: pricing · anthropic-claude-sonnet-5: pricing · anthropic-claude-opus-5-5: pricing · anthropic-claude-fable-5-1: pricing · anthropic-claude-sonnet-5-5: pricing · anthropic-claude-haiku-5-5: pricing · Retrieved 2026-10-28T09:19:29.129Z · Effective: Unknown · Checked: 2026-10-28T09:19:29.129Z · Reviewed: Unknown · Review after: 2026-11-07T09:19:29.129Z · Freshness: stale · Confidence: A

• Raw SHA-256: 31ba8ce4d0f3b4b6a4e96e53e6edb58ccd3cb1ccea5916576e0e230549b72670; collector anthropic-markdown-2.
• Pricing observation only. Display-name mapping is not official API identity evidence; official IDs carry their own model-page identity records (#449). No new access, availability, weight, retirement or model metadata assertion.
• This page provides detailed pricing information for Anthropic's models and features. All prices are in USD. For the most current pricing information, visit claude.com/pricing.
• The following table shows pricing for all Claude models: 1 Cache hits and refreshes on Claude Fable 5.1 and Claude Mythos 5.1 are priced at 0.025x the base input price. 2 Cache hits and refreshes on Claude Opus 5.5 and Claude Sonnet 5.5 are priced at 0.05x the base input price. All other models use the standard 0.1x multiplier. 3 The $2/$10 per million input/output token pricing for Claude Sonnet 5, announced at launch as introductory pricing through August 31, 2026, is now the standard price. The previously scheduled increase to $3/$15 per million input/output tokens on September 1, 2026 will not occur. MTok: Million tokens. $5 / MTok is $5 for every million tokens. 5m cache writes: Writing a prompt prefix to the 5-minute prompt cache. 1h cache writes: Writing a prompt prefix to the 1-hour prompt cache. Cache hits and refreshes: Reading a prompt prefix from the prompt cache, which also refreshes it. Limited access: Available only to organizations verified through Anthropic’s verification programs, such as the Cyber Verification Program. To request access, apply to the program that covers your use case. Retired: May still be available on other cloud platforms. See Model deprecations for more. Claude 4.7 and later models and Claude Mythos Preview use a newer tokenizer that contributes to their improved performance on a wide range of tasks. This tokenizer produces approximately 30% more tokens for the same text. The exact increase depends on the content and workload shape. Claude Sonnet 4.6 and earlier models use the previous tokenizer.
• Prompt caching reduces costs and latency by reusing previously processed portions of your prompt across API calls. Instead of reprocessing the same large system prompt, document, or conversation history on every request, the API reads from cache at a fraction of the standard input price. There are two ways to enable prompt caching: Automatic caching: Add a single cache_control field at the top level of your request. The system automatically manages cache breakpoints as conversations grow. This is the recommended starting point for most use cases. Explicit cache breakpoints: Place cache_control directly on individual content blocks for fine-grained control over exactly what gets cached. Prompt caching uses the following pricing multipliers relative to base input token rates: Cache operation Multiplier Duration 5-minute cache write 1.25x base input price Cache valid for 5 minutes 1-hour cache write 2x base input price Cache valid for 1 hour Cache read (hit) 0.1x base input price (0.025x on Claude Fable 5.1 and Claude Mythos 5.1; 0.05x on Claude Opus 5.5 and Claude Sonnet 5.5) Same duration as the preceding write Cache write tokens are charged when content is first stored. Cache read tokens are charged when a subsequent request retrieves the cached content. A cache hit costs 10% of the standard input price, which means caching pays off after one cache read for the 5-minute duration (1.25x write), or after two cache reads for the 1-hour duration (2x write). On Claude Fable 5.1 and Claude Mythos 5.1, a cache hit costs 2.5% of the standard input price ($0.25 USD per million tokens). On Claude Opus 5.5 and Claude Sonnet 5.5, a cache hit costs 5% of the standard input price ($0.20 USD per million tokens on Claude Opus 5.5, $0.10 USD on Claude Sonnet 5.5). These multipliers stack with other pricing modifiers, including the Batch API discount and data residency. For implementation details, supported models, and code examples, see Prompt caching.
• For Claude 4.6 and later models, specifying US-only inference through the inference_geo parameter incurs a 1.1x multiplier on all token pricing categories, including input tokens, output tokens, cache writes, and cache reads. On Claude Haiku 5.5, the multiplier also applies to the higher prices for prompts over 100,000 tokens (see Long context pricing). Global routing (the default) uses standard pricing. This applies to the Claude API (first-party), Claude Platform on AWS, and Claude Models on Google Cloud (Anthropic managed). On Claude in Microsoft Foundry, the same 1.1x multiplier applies to deployments that use the US Data Zone Standard deployment type (see Inference geography). Partner-operated platforms, Bedrock and Claude Models on Google Cloud (Google managed), have independent regional pricing. See Bedrock pricing and Google Cloud pricing for details. Earlier models do not support the inference_geo parameter and always use standard pricing; requests that include the parameter on these models return a 400 error. For more information, see Data residency.
• Claude 4.6 and later models (except Claude Haiku 5.5) and Claude Mythos Preview include the full 1M token context window at standard pricing. (A 900k-token request is billed at the same per-token rate as a 9k-token request.) Prompt caching and batch processing discounts apply at standard rates across the full context window. Claude Haiku 5.5 is priced by prompt length: a request whose prompt is over 100,000 tokens pays higher prices. A request's prompt length counts all of its input tokens, including cache reads and cache writes. Each request is priced on its own: a request over the threshold pays the higher prices even when part of its prompt is a cache hit, and earlier requests keep the prices they were billed at. The model pricing and batch processing pricing sections list both sets of prices.

Primary source ↗ · vendor-docs · Claims: anthropic-claude-fable-5: pricing · anthropic-claude-haiku-4-5: pricing · anthropic-claude-opus-5: pricing · anthropic-claude-sonnet-5: pricing · anthropic-claude-opus-5-5: pricing · anthropic-claude-fable-5-1: pricing · anthropic-claude-sonnet-5-5: pricing · anthropic-claude-haiku-5-5: pricing · Retrieved 2026-10-30T04:28:58.251Z · Effective: Unknown · Checked: 2026-10-30T04:28:58.251Z · Reviewed: Unknown · Review after: 2026-11-09T04:28:58.251Z · Freshness: fresh · Confidence: A

• Raw SHA-256: 79f8d09b6562f37f5edba5108b871cdd13dba26603efbd748e77a40f18c5ff53; collector anthropic-markdown-2.
• Pricing observation only. Display-name mapping is not official API identity evidence; official IDs carry their own model-page identity records (#449). No new access, availability, weight, retirement or model metadata assertion.
• This page provides detailed pricing information for Anthropic's models and features. All prices are in USD. For the most current pricing information, visit claude.com/pricing.
• The following table shows pricing for all Claude models: 1 Cache hits and refreshes on Claude Fable 5.1 and Claude Mythos 5.1 are priced at 0.025x the base input price. 2 Cache hits and refreshes on Claude Opus 5.5 and Claude Sonnet 5.5 are priced at 0.05x the base input price. All other models use the standard 0.1x multiplier. 3 The $2/$10 per million input/output token pricing for Claude Sonnet 5, announced at launch as introductory pricing through August 31, 2026, is now the standard price. The previously scheduled increase to $3/$15 per million input/output tokens on September 1, 2026 will not occur. MTok: Million tokens. $5 / MTok is $5 for every million tokens. 5m cache writes: Writing a prompt prefix to the 5-minute prompt cache. 1h cache writes: Writing a prompt prefix to the 1-hour prompt cache. Cache hits and refreshes: Reading a prompt prefix from the prompt cache, which also refreshes it. Limited access: Available only to organizations verified through Anthropic’s verification programs, such as the Cyber Verification Program. To request access, apply to the program that covers your use case. Retired: May still be available on other cloud platforms. See Model deprecations for more. Claude 4.7 and later models and Claude Mythos Preview use a newer tokenizer that contributes to their improved performance on a wide range of tasks. This tokenizer produces approximately 30% more tokens for the same text. The exact increase depends on the content and workload shape. Claude Sonnet 4.6 and earlier models use the previous tokenizer.
• Prompt caching reduces costs and latency by reusing previously processed portions of your prompt across API calls. Instead of reprocessing the same large system prompt, document, or conversation history on every request, the API reads from cache at a fraction of the standard input price. There are two ways to enable prompt caching: Automatic caching: Add a single cache_control field at the top level of your request. The system automatically manages cache breakpoints as conversations grow. This is the recommended starting point for most use cases. Explicit cache breakpoints: Place cache_control directly on individual content blocks for fine-grained control over exactly what gets cached. Prompt caching uses the following pricing multipliers relative to base input token rates: Cache operation Multiplier Duration 5-minute cache write 1.25x base input price Cache valid for 5 minutes 1-hour cache write 2x base input price Cache valid for 1 hour Cache read (hit) 0.1x base input price (0.025x on Claude Fable 5.1 and Claude Mythos 5.1; 0.05x on Claude Opus 5.5 and Claude Sonnet 5.5) Same duration as the preceding write Cache write tokens are charged when content is first stored. Cache read tokens are charged when a subsequent request retrieves the cached content. A cache hit costs 10% of the standard input price, which means caching pays off after one cache read for the 5-minute duration (1.25x write), or after two cache reads for the 1-hour duration (2x write). On Claude Fable 5.1 and Claude Mythos 5.1, a cache hit costs 2.5% of the standard input price ($0.25 USD per million tokens). On Claude Opus 5.5 and Claude Sonnet 5.5, a cache hit costs 5% of the standard input price ($0.20 USD per million tokens on Claude Opus 5.5, $0.10 USD on Claude Sonnet 5.5). These multipliers stack with other pricing modifiers, including the Batch API discount and data residency. For implementation details, supported models, and code examples, see Prompt caching.
• For Claude 4.6 and later models, specifying US-only inference through the inference_geo parameter incurs a 1.1x multiplier on all token pricing categories, including input tokens, output tokens, cache writes, and cache reads. On Claude Haiku 5.5, the multiplier also applies to the higher prices for prompts over 100,000 tokens (see Long context pricing). Global routing (the default) uses standard pricing. This applies to the Claude API (first-party), Claude Platform on AWS, and Claude Models on Google Cloud (Anthropic managed). On Claude in Microsoft Foundry, the same 1.1x multiplier applies to deployments that use the US Data Zone Standard deployment type (see Inference geography). Partner-operated platforms, Bedrock and Claude Models on Google Cloud (Google managed), have independent regional pricing. See Bedrock pricing and Google Cloud pricing for details. Earlier models do not support the inference_geo parameter and always use standard pricing; requests that include the parameter on these models return a 400 error. For more information, see Data residency.
• Claude 4.6 and later models (except Claude Haiku 5.5) and Claude Mythos Preview include the full 1M token context window at standard pricing. (A 900k-token request is billed at the same per-token rate as a 9k-token request.) Prompt caching and batch processing discounts apply at standard rates across the full context window. Claude Haiku 5.5 is priced by prompt length: a request whose prompt is over 100,000 tokens pays higher prices. A request's prompt length counts all of its input tokens, including cache reads and cache writes. Each request is priced on its own: a request over the threshold pays the higher prices even when part of its prompt is a cache hit, and earlier requests keep the prices they were billed at. The model pricing and batch processing pricing sections list both sets of prices.