compute.info

Compute prices, compared.

Cloud VMs, AI APIs and hardware, pulled from official sources and dated.

  • 2,158VM prices
  • 11cloud providers
  • 219regions
  • 80AI API offers
  • 16AI providers
  • 9hardware items
Cheapest 2 vCPU VM · billed in USDOn-demand · 4–8 GiB · per month (vendor cap or 730 × hourly)
Oracle Cloud InfrastructureOracle Cloud Infrastructure: $16.06 · VM.Standard.A2.Flex (1 OCPU, 4 GB) · 4 GiB · Oracle public price list; deployment region not specified$16.06DigitalOceanDigitalOcean: $24.00 · Basic (s-2vcpu-4gb) · 4 GiB · Amsterdam 3$24.00AWSAWS: $24.53 · t4g.medium · 4 GiB · US East (N. Virginia)$24.53Akamai Connected CloudAkamai Connected Cloud: $36.00 · Dedicated 4GB · 4 GiB · Tokyo 2, JP$36.00Google CloudGoogle Cloud: $43.75 · c2d-standard-2 · 8 GiB · asia-south1$43.75VultrVultr: $43.80 · vx1-g-2c-8g · 8 GiB · Amsterdam (ams)$43.80
Cheapest 2 vCPU VM · billed in USD
ItemValueDetail
Oracle Cloud Infrastructure$16.06VM.Standard.A2.Flex (1 OCPU, 4 GB) · 4 GiB · Oracle public price list; deployment region not specified
DigitalOcean$24.00Basic (s-2vcpu-4gb) · 4 GiB · Amsterdam 3
AWS$24.53t4g.medium · 4 GiB · US East (N. Virginia)
Akamai Connected Cloud$36.00Dedicated 4GB · 4 GiB · Tokyo 2, JP
Google Cloud$43.75c2d-standard-2 · 8 GiB · asia-south1
Vultr$43.80vx1-g-2c-8g · 8 GiB · Amsterdam (ams)

Shared, dedicated and Arm instances mixed; not a performance comparison.

Cheapest 2 vCPU VM · billed in EUROn-demand · 4–8 GiB · per month (vendor cap or 730 × hourly) · not converted
Hetzner CloudHetzner Cloud: €5.99 · CX23 · 4 GiB · Falkenstein (FSN1)€5.99UpCloudUpCloud: €12.00 · STARTER-2xCPU-4GB · 4 GiB · Sydney #1 (au-syd1)€12.00ScalewayScaleway: €16.79 · BASIC2-A2C-4G · 4 GiB · Paris 1 (fr-par-1)€16.79
Cheapest 2 vCPU VM · billed in EUR
ItemValueDetail
Hetzner Cloud€5.99CX23 · 4 GiB · Falkenstein (FSN1)
UpCloud€12.00STARTER-2xCPU-4GB · 4 GiB · Sydney #1 (au-syd1)
Scaleway€16.79BASIC2-A2C-4G · 4 GiB · Paris 1 (fr-par-1)

Excluding tax: Hetzner Cloud, Scaleway. Tax treatment not stated: UpCloud. Not like-for-like across groups.

AI API input pricesUSD per 1M input tokens · headline rates, conditions labeled · log scale
$0.01$0.10$1$10Fireworks AIFireworks AI · NVIDIA Nemotron 3.5 Lightning 30B A3B: $0.05Fireworks AI · GLM 5.3 Flash: $0.15Fireworks AI · OpenAI GPT OSS 120B: $0.15Fireworks AI · GLM 5.3 Flash (US): $0.225Fireworks AI · DeepSeek V4.1 Flash: $0.30Fireworks AI · MiniMax M3: $0.30Fireworks AI · DeepSeek V4.1 Flash (US): $0.45Fireworks AI · NVIDIA Nemotron 3 Ultra (Preview): $0.60Fireworks AI · GLM 5.3: $1.40Fireworks AI · Qwen 3.8 Max: $2Fireworks AI · GLM 5.3 (US): $2.10Fireworks AI · Ember-1: $3Fireworks AI · Kimi K3: $3Fireworks AI · Kimi K3 (US): $4.50GoogleGoogle · Gemini 2.5 Flash-Lite (text/image/video): $0.10Google · Gemini 3.5 Flash-Lite (text/image/video/audio input): $0.30Google · Gemini 3.7 Flash: $0.75Google · Gemini 2.5 Pro (≤200K prompt): $1.25Google · Gemini 3.1 Pro Preview (≤200K prompt): $2Mistral AIMistral AI · Ministral 3 3B: $0.10Mistral AI · Mistral Small 4: $0.15Mistral AI · Ministral 3 8B: $0.15Mistral AI · Ministral 3 14B: $0.20Mistral AI · Codestral: $0.30Mistral AI · Mistral Large 3: $0.50Mistral AI · Mistral Large 4: $0.68Mistral AI · Mistral Medium 3.5: $1.50OpenAIOpenAI · GPT-6 Luna (short context; availability unverified): $0.10OpenAI · GPT-5.4 nano (short context): $0.20OpenAI · GPT-5.6 Luna (short context): $0.20OpenAI · GPT-5.4 mini (short context): $0.75OpenAI · GPT-5.6 Terra (short context): $2OpenAI · GPT-6 Sol (short context; availability unverified): $2OpenAI · GPT-6.1 Sol (short context; availability unverified): $2OpenAI · GPT-5.6 Sol (short context): $4OpenAI · GPT-6 Astra (short context; availability unverified): $10OpenRouterOpenRouter · Meta: Llama 3.3 70B Instruct (routed listing): $0.10OpenRouter · DeepSeek: DeepSeek V3.1 (routed listing): $0.25OpenRouter · MiniMax: MiniMax M2 (routed listing): $0.30StepFunStepFun · Step-3.5-Flash: $0.10StepFun · Step-3.7-Flash: $0.20StepFun · Step-5 Preview: $1Novita AINovita AI · Llama 3.3 70B Instruct (Serverless tier; availability unverified): $0.135Novita AI · MiniMax M2 (Serverless tier; availability unverified): $0.30Novita AI · GLM 5 (Serverless tier; availability unverified): $1Xiaomi MiMoXiaomi MiMo · mimo-v2.6-flash: $0.14Xiaomi MiMo · mimo-v2.6-pro: $0.435Xiaomi MiMo · mimo-v2.6-pro-ultraspeed: $4.35Alibaba Cloud Model StudioAlibaba Cloud Model Studio · Qwen3.8-Flash (≤1M tokens): $0.15Alibaba Cloud Model Studio · Qwen3.7-Plus (≤256K tokens): $0.40Alibaba Cloud Model Studio · Qwen3.8-Max (≤1M tokens): $2Z.aiZ.ai · GLM-5.3-Flash: $0.15Z.ai · GLM-5.3-FlashX: $0.37Z.ai · GLM-5.3: $1.40Z.ai · GLM-5.2: $1.40DeepSeekDeepSeek · DeepSeek V4.1 Flash: $0.30DeepSeek · DeepSeek V4 Pro 0813: $1.32MiniMaxMiniMax · MiniMax-M3 (≤512K context): $0.30MiniMax · MiniMax-M2.7: $0.30MiniMax · MiniMax-M2.7-highspeed: $0.60Moonshot AIMoonshot AI · Kimi K2.6: $0.95Moonshot AI · Kimi K2.7 Code: $0.95Moonshot AI · Kimi K2.7 Code Highspeed (Highspeed tier): $1.90Moonshot AI · Kimi K3: $3AnthropicAnthropic · Claude Haiku 4.5: $1Anthropic · Claude Sonnet 5 (≤1M tokens): $2Anthropic · Claude Opus 5.5 (≤1M tokens; availability unverified): $4Anthropic · Claude Opus 5 (≤1M tokens): $5Anthropic · Claude Fable 5 (≤1M tokens): $10Anthropic · Claude Fable 5.1 (≤1M tokens; availability unverified): $10xAIxAI · grok-build-0.1 (<200K tokens): $1xAI · grok-4.3 (<200K tokens): $1.25xAI · grok-4.20-0309-reasoning (<200K tokens): $1.25xAI · grok-4.20-0309-non-reasoning (<200K tokens): $1.25xAI · grok-4.20-multi-agent-0309 (<200K tokens): $1.25xAI · grok-4.7 (<200K tokens): $2xAI · grok-4.6 (<200K tokens): $2xAI · grok-4.5 (<200K tokens): $2
AI API input prices
CategoryLowest valueLowest-pricedHighest valueItems
Fireworks AI$0.05NVIDIA Nemotron 3.5 Lightning 30B A3B$4.5014
Google$0.10Gemini 2.5 Flash-Lite (text/image/video)$25
Mistral AI$0.10Ministral 3 3B$1.508
OpenAI$0.10GPT-6 Luna (short context; availability unverified)$109
OpenRouter$0.10Meta: Llama 3.3 70B Instruct (routed listing)$0.303
StepFun$0.10Step-3.5-Flash$13
Novita AI$0.135Llama 3.3 70B Instruct (Serverless tier; availability unverified)$13
Xiaomi MiMo$0.14mimo-v2.6-flash$4.353
Alibaba Cloud Model Studio$0.15Qwen3.8-Flash (≤1M tokens)$23
Z.ai$0.15GLM-5.3-Flash$1.404
DeepSeek$0.30DeepSeek V4.1 Flash$1.322
MiniMax$0.30MiniMax-M3 (≤512K context), MiniMax-M2.7$0.603
Moonshot AI$0.95Kimi K2.6, Kimi K2.7 Code$34
Anthropic$1Claude Haiku 4.5$106
xAI$1grok-build-0.1 (<200K tokens)$28

2 other offers have non-USD, multimodal or ambiguous input rates and are listed on the AI page. Includes promotional or time-limited prices: Gemini 3.7 Flash (Google) until 2026-12-31; GPT-5.6 Sol (OpenAI) promotional, at least through 2026-11-21; Mistral Large 4 (Mistral AI) sale price (original input price USD 1.36 / 1000000 tokens), no end date shown.

Cloud virtual machines

Compare VM compute and list prices

Google Cloudc2d-standard-2 · Default USD price scope2 vCPU8 GiB memory≈$66.28 / mo730 hour estimateHetzner CloudCCX13 · Falkenstein (FSN1)2 vCPU8 GiB memory€43.49 / movendor capOVHcloudVPS-1 · Ireland storefront; deployment region not specified2 vCPU4 GiB memory€3.81 / movendor priceScalewayBASIC2-A2C-4G · Paris (fr-par-1)2 vCPU4 GiB memory≈€16.79 / mo730 hour estimate
Browse all 2158 VM offerings

Start with a requirement: compare VMs with 32 GiB RAM for memory capacity, VMs under $10 per month for a base-compute budget, or Hetzner and DigitalOcean at matched memory sizes for a provider shortlist. Check storage, traffic and tax separately.

AI

API services and models you can run yourself

Claude Fable 5Anthropic · API offerUSD 10 / 1000000 tokensinput · see source conditionsUSD 50 / 1000000 tokensoutput · retrieved 2026-09-29Gemini 2.5 Flash-Lite (text/image/video)Google · API offerUSD 0.10 / 1000000 tokensinput · see source conditionsUSD 0.40 / 1000000 tokensoutput · retrieved 2026-10-09Kimi K2.6Moonshot AI · API offerUSD 0.95 / 1000000 tokensinput · see source conditionsUSD 4.00 / 1000000 tokensoutput · retrieved 2026-09-30GPT-5.4 miniOpenAI · API offerUSD 0.75 / 1000000 tokens / Unknown / 1000000 tokensinput · see source conditionsUSD 4.50 / 1000000 tokens / Unknown / 1000000 tokensoutput · retrieved 2026-10-09DeepSeek V4.1 FlashDeepSeek · API offerUSD 0.3 / 1000000 tokens / USD 0.15 / 1000000 tokensinput · see source conditionsUSD 1.2 / 1000000 tokens / USD 0.6 / 1000000 tokensoutput · retrieved 2026-09-30DeepSeek R1Downloadable · MIT671Bparameters503.3 GBest. Q4 weightsDeepSeek V3Downloadable · DeepSeek License Agreement v1.0 (use restrictions)671Bparameters503.3 GBest. Q4 weights
Browse 80 API offers and 10 downloadable model records

Hardware

GPUs and unified-memory systems

NVIDIA GeForce RTX 4090consumer gpu24 GBdedicatedGDDR6Xmemory typeAMD Radeon RX 7900 XTXconsumer gpu24 GBdedicatedGDDR6memory typeApple Mac Studio M3 Ultra (maximum configuration)unified memory system512 GBunified / sharedUnified memorymemory typeNVIDIA GeForce RTX 5090consumer gpu32 GBdedicatedGDDR7memory typeAMD Radeon PRO W7900workstation gpu48 GBdedicatedGDDR6 ECCmemory type
Browse all 9 hardware records
Every price links to its source and retrieval date.Methodology →Sources →