{
 "checkedAt": "2026-10-18",
 "currency": "USD",
 "unit": "per 1M tokens",
 "notes": "Standard (non-batch, non-priority, global-endpoint) list prices. Optional fields: longContext.cachedInput, scheduledChange (announced future price), offPeak (DeepSeek), regularPrice (pre-sale price).",
 "providers": {
  "anthropic": {
   "name": "Anthropic",
   "pricingUrl": "https://platform.claude.com/docs/en/about-claude/pricing"
  },
  "openai": {
   "name": "OpenAI",
   "pricingUrl": "https://developers.openai.com/api/docs/pricing"
  },
  "google": {
   "name": "Google (Gemini API)",
   "pricingUrl": "https://ai.google.dev/gemini-api/docs/pricing"
  },
  "xai": {
   "name": "xAI",
   "pricingUrl": "https://docs.x.ai/developers/pricing"
  },
  "deepseek": {
   "name": "DeepSeek",
   "pricingUrl": "https://api-docs.deepseek.com/quick_start/pricing"
  },
  "mistral": {
   "name": "Mistral AI",
   "pricingUrl": "https://docs.mistral.ai/inference/pricing"
  }
 },
 "models": [
  {
   "id": "claude-fable-5-1",
   "provider": "anthropic",
   "name": "Claude Fable 5.1",
   "tier": "flagship",
   "input": 10,
   "output": 50,
   "cachedInput": 0.25,
   "cacheWrite": 12.5,
   "batchDiscount": 0.5,
   "contextWindow": 1000000,
   "maxOutput": 128000,
   "longContext": null,
   "released": "2026-09",
   "source": "https://platform.claude.com/docs/en/about-claude/pricing",
   "verifiedAt": "2026-10-18",
   "confidence": "official",
   "notes": "Top model for demanding reasoning / long-horizon agents. Cache reads 0.025x input. Adaptive thinking always on; thinking billed as output. 1M context at standard price (no long-context surcharge). 1h cache write = 2x input. Uses newer tokenizer (~30% more tokens than Claude 4.6 and earlier for same text). US-only inference_geo = 1.1x."
  },
  {
   "id": "claude-opus-5-5",
   "provider": "anthropic",
   "name": "Claude Opus 5.5",
   "tier": "flagship",
   "input": 4,
   "output": 20,
   "cachedInput": 0.2,
   "cacheWrite": 5,
   "batchDiscount": 0.5,
   "contextWindow": 1000000,
   "maxOutput": 128000,
   "longContext": null,
   "released": "2026-09",
   "source": "https://platform.claude.com/docs/en/about-claude/pricing",
   "verifiedAt": "2026-10-18",
   "confidence": "official",
   "notes": "Anthropic's recommended default. Cache reads 0.05x input. Fast mode $8/$40. Thinking billed as output. 1M context at standard price (no long-context surcharge). 1h cache write = 2x input. Uses newer tokenizer (~30% more tokens than Claude 4.6 and earlier for same text). US-only inference_geo = 1.1x."
  },
  {
   "id": "claude-sonnet-5-5",
   "provider": "anthropic",
   "name": "Claude Sonnet 5.5",
   "tier": "mid",
   "input": 2,
   "output": 10,
   "cachedInput": 0.1,
   "cacheWrite": 2.5,
   "batchDiscount": 0.5,
   "contextWindow": 1000000,
   "maxOutput": 128000,
   "longContext": null,
   "released": "2026-09",
   "source": "https://platform.claude.com/docs/en/about-claude/pricing",
   "verifiedAt": "2026-10-18",
   "confidence": "official",
   "notes": "Cache reads 0.05x input. Thinking billed as output. 1M context at standard price (no long-context surcharge). 1h cache write = 2x input. Uses newer tokenizer (~30% more tokens than Claude 4.6 and earlier for same text). US-only inference_geo = 1.1x."
  },
  {
   "id": "claude-haiku-5-5",
   "provider": "anthropic",
   "name": "Claude Haiku 5.5",
   "tier": "small",
   "input": 0.1,
   "output": 0.5,
   "cachedInput": 0.01,
   "cacheWrite": 0.125,
   "batchDiscount": 0.5,
   "contextWindow": 1000000,
   "maxOutput": 128000,
   "longContext": {
    "threshold": 100000,
    "input": 0.5,
    "output": 2.5,
    "cachedInput": 0.05,
    "cacheWrite": 0.625
   },
   "released": "2026-10",
   "source": "https://platform.claude.com/docs/en/about-claude/pricing",
   "verifiedAt": "2026-10-18",
   "confidence": "official",
   "notes": "Priced by prompt length: prompts over 100K tokens (incl. cache reads/writes) pay 5x rates for the whole request. Batch: $0.05/$0.25 (<=100K), $0.25/$1.25 (>100K)."
  },
  {
   "id": "claude-opus-5",
   "provider": "anthropic",
   "name": "Claude Opus 5",
   "tier": "flagship",
   "input": 5,
   "output": 25,
   "cachedInput": 0.5,
   "cacheWrite": 6.25,
   "batchDiscount": 0.5,
   "contextWindow": 1000000,
   "maxOutput": 128000,
   "longContext": null,
   "released": "2026-07",
   "source": "https://platform.claude.com/docs/en/about-claude/pricing",
   "verifiedAt": "2026-10-18",
   "confidence": "official",
   "notes": "Legacy (superseded by Opus 5.5, which is cheaper). 1M context at standard price (no long-context surcharge). 1h cache write = 2x input. Uses newer tokenizer (~30% more tokens than Claude 4.6 and earlier for same text). US-only inference_geo = 1.1x."
  },
  {
   "id": "claude-sonnet-5",
   "provider": "anthropic",
   "name": "Claude Sonnet 5",
   "tier": "mid",
   "input": 2,
   "output": 10,
   "cachedInput": 0.2,
   "cacheWrite": 2.5,
   "batchDiscount": 0.5,
   "contextWindow": 1000000,
   "maxOutput": 128000,
   "longContext": null,
   "released": "2026-06",
   "source": "https://platform.claude.com/docs/en/about-claude/pricing",
   "verifiedAt": "2026-10-18",
   "confidence": "official",
   "notes": "Legacy. Launch \"introductory\" $2/$10 pricing made permanent; planned Sept 1, 2026 increase to $3/$15 cancelled. 1M context at standard price (no long-context surcharge). 1h cache write = 2x input. Uses newer tokenizer (~30% more tokens than Claude 4.6 and earlier for same text). US-only inference_geo = 1.1x."
  },
  {
   "id": "claude-opus-4-8",
   "provider": "anthropic",
   "name": "Claude Opus 4.8",
   "tier": "flagship",
   "input": 5,
   "output": 25,
   "cachedInput": 0.5,
   "cacheWrite": 6.25,
   "batchDiscount": 0.5,
   "contextWindow": 1000000,
   "maxOutput": 128000,
   "longContext": null,
   "released": "2026-05",
   "source": "https://platform.claude.com/docs/en/about-claude/pricing",
   "verifiedAt": "2026-10-18",
   "confidence": "official",
   "notes": "Legacy. 1M context at standard price (no long-context surcharge). 1h cache write = 2x input. Uses newer tokenizer (~30% more tokens than Claude 4.6 and earlier for same text). US-only inference_geo = 1.1x."
  },
  {
   "id": "claude-sonnet-4-6",
   "provider": "anthropic",
   "name": "Claude Sonnet 4.6",
   "tier": "mid",
   "input": 3,
   "output": 15,
   "cachedInput": 0.3,
   "cacheWrite": 3.75,
   "batchDiscount": 0.5,
   "contextWindow": 1000000,
   "maxOutput": 128000,
   "longContext": null,
   "released": "2026-02",
   "source": "https://platform.claude.com/docs/en/about-claude/pricing",
   "verifiedAt": "2026-10-18",
   "confidence": "official",
   "notes": "Legacy but widely used. Previous tokenizer. 1M context at standard price."
  },
  {
   "id": "claude-haiku-4-5-20251001",
   "provider": "anthropic",
   "name": "Claude Haiku 4.5",
   "tier": "small",
   "input": 1,
   "output": 5,
   "cachedInput": 0.1,
   "cacheWrite": 1.25,
   "batchDiscount": 0.5,
   "contextWindow": 200000,
   "maxOutput": 64000,
   "longContext": null,
   "released": "2025-10",
   "source": "https://platform.claude.com/docs/en/about-claude/pricing",
   "verifiedAt": "2026-10-18",
   "confidence": "official",
   "notes": "Legacy; alias claude-haiku-4-5. Now 10x the price of Haiku 5.5. Retirement \"not sooner than Oct 15, 2026\" - check deprecations page."
  },
  {
   "id": "gpt-6-astra",
   "provider": "openai",
   "name": "GPT-6 Astra",
   "tier": "flagship",
   "input": 10,
   "output": 50,
   "cachedInput": 1,
   "cacheWrite": 12.5,
   "batchDiscount": 0.5,
   "contextWindow": 1050000,
   "maxOutput": 128000,
   "longContext": {
    "threshold": 272000,
    "input": 20,
    "output": 75,
    "cachedInput": 2
   },
   "released": null,
   "source": "https://developers.openai.com/api/docs/pricing",
   "verifiedAt": "2026-10-18",
   "confidence": "official",
   "notes": "OpenAI flagship. Max input 922K. Fast mode 2x. Long context: prompts >272K input tokens bill the whole request at 2x input/cache and 1.5x output. Reasoning tokens billed as output. Regional/data-residency endpoints +10%."
  },
  {
   "id": "gpt-6.1-sol",
   "provider": "openai",
   "name": "GPT-6.1 Sol",
   "tier": "mid",
   "input": 2,
   "output": 10,
   "cachedInput": 0.1,
   "cacheWrite": 2.5,
   "batchDiscount": 0.5,
   "contextWindow": 1050000,
   "maxOutput": 128000,
   "longContext": {
    "threshold": 272000,
    "input": 4,
    "output": 15,
    "cachedInput": 0.2
   },
   "released": null,
   "source": "https://developers.openai.com/api/docs/pricing",
   "verifiedAt": "2026-10-18",
   "confidence": "official",
   "notes": "Mid tier, near-Astra quality. Cached input is 0.05x (vs 0.1x for gpt-6-sol). Long context: prompts >272K input tokens bill the whole request at 2x input/cache and 1.5x output. Reasoning tokens billed as output. Regional/data-residency endpoints +10%."
  },
  {
   "id": "gpt-6-luna",
   "provider": "openai",
   "name": "GPT-6 Luna",
   "tier": "small",
   "input": 0.1,
   "output": 0.5,
   "cachedInput": 0.01,
   "cacheWrite": 0.125,
   "batchDiscount": 0.5,
   "contextWindow": 1050000,
   "maxOutput": 128000,
   "longContext": {
    "threshold": 272000,
    "input": 0.2,
    "output": 0.75,
    "cachedInput": 0.02
   },
   "released": null,
   "source": "https://developers.openai.com/api/docs/pricing",
   "verifiedAt": "2026-10-18",
   "confidence": "official",
   "notes": "Small/efficient tier. Long context: prompts >272K input tokens bill the whole request at 2x input/cache and 1.5x output. Reasoning tokens billed as output. Regional/data-residency endpoints +10%."
  },
  {
   "id": "gpt-6-sol",
   "provider": "openai",
   "name": "GPT-6 Sol",
   "tier": "mid",
   "input": 2,
   "output": 10,
   "cachedInput": 0.2,
   "cacheWrite": 2.5,
   "batchDiscount": 0.5,
   "contextWindow": 1050000,
   "maxOutput": 128000,
   "longContext": {
    "threshold": 272000,
    "input": 4,
    "output": 15,
    "cachedInput": 0.4
   },
   "released": null,
   "source": "https://developers.openai.com/api/docs/pricing",
   "verifiedAt": "2026-10-18",
   "confidence": "official",
   "notes": "Superseded by gpt-6.1-sol (same price, cheaper cache reads). Long context: prompts >272K input tokens bill the whole request at 2x input/cache and 1.5x output. Reasoning tokens billed as output. Regional/data-residency endpoints +10%."
  },
  {
   "id": "gpt-5.6-sol",
   "provider": "openai",
   "name": "GPT-5.6 Sol",
   "tier": "flagship",
   "input": 4,
   "output": 20,
   "cachedInput": 0.4,
   "cacheWrite": 5,
   "batchDiscount": 0.5,
   "contextWindow": 1050000,
   "maxOutput": 128000,
   "longContext": {
    "threshold": 272000,
    "input": 8,
    "output": 30,
    "cachedInput": 0.8
   },
   "released": null,
   "source": "https://developers.openai.com/api/docs/pricing",
   "verifiedAt": "2026-10-18",
   "confidence": "official",
   "notes": "Previous generation (alias gpt-5.6). Page says \"promotional pricing is available at least through November 21, 2026\" - may rise after. Long context: prompts >272K input tokens bill the whole request at 2x input/cache and 1.5x output. Reasoning tokens billed as output. Regional/data-residency endpoints +10%."
  },
  {
   "id": "gpt-5.6-terra",
   "provider": "openai",
   "name": "GPT-5.6 Terra",
   "tier": "mid",
   "input": 2,
   "output": 12,
   "cachedInput": 0.2,
   "cacheWrite": 2.5,
   "batchDiscount": 0.5,
   "contextWindow": 1050000,
   "maxOutput": 128000,
   "longContext": {
    "threshold": 272000,
    "input": 4,
    "output": 18,
    "cachedInput": 0.4
   },
   "released": null,
   "source": "https://developers.openai.com/api/docs/pricing",
   "verifiedAt": "2026-10-18",
   "confidence": "official",
   "notes": "Previous generation. Long context: prompts >272K input tokens bill the whole request at 2x input/cache and 1.5x output. Reasoning tokens billed as output. Regional/data-residency endpoints +10%."
  },
  {
   "id": "gpt-5.6-luna",
   "provider": "openai",
   "name": "GPT-5.6 Luna",
   "tier": "small",
   "input": 0.2,
   "output": 1.2,
   "cachedInput": 0.02,
   "cacheWrite": 0.25,
   "batchDiscount": 0.5,
   "contextWindow": 1050000,
   "maxOutput": 128000,
   "longContext": {
    "threshold": 272000,
    "input": 0.4,
    "output": 1.8,
    "cachedInput": 0.04
   },
   "released": null,
   "source": "https://developers.openai.com/api/docs/pricing",
   "verifiedAt": "2026-10-18",
   "confidence": "official",
   "notes": "Previous generation. Long context: prompts >272K input tokens bill the whole request at 2x input/cache and 1.5x output. Reasoning tokens billed as output. Regional/data-residency endpoints +10%."
  },
  {
   "id": "gpt-5.5",
   "provider": "openai",
   "name": "GPT-5.5",
   "tier": "flagship",
   "input": 5,
   "output": 30,
   "cachedInput": 0.5,
   "cacheWrite": null,
   "batchDiscount": 0.5,
   "contextWindow": 1050000,
   "maxOutput": 128000,
   "longContext": {
    "threshold": 272000,
    "input": 10,
    "output": 45,
    "cachedInput": 1
   },
   "released": "2026-04",
   "source": "https://developers.openai.com/api/docs/pricing",
   "verifiedAt": "2026-10-18",
   "confidence": "official",
   "notes": "Snapshot gpt-5.5-2026-04-23. No separate cache-write charge. Long context: prompts >272K input tokens bill the whole request at 2x input/cache and 1.5x output. Reasoning tokens billed as output. Regional/data-residency endpoints +10%."
  },
  {
   "id": "gpt-5.4-mini",
   "provider": "openai",
   "name": "GPT-5.4 mini",
   "tier": "small",
   "input": 0.75,
   "output": 4.5,
   "cachedInput": 0.075,
   "cacheWrite": null,
   "batchDiscount": 0.5,
   "contextWindow": 400000,
   "maxOutput": 128000,
   "longContext": null,
   "released": "2026-03",
   "source": "https://developers.openai.com/api/docs/pricing",
   "verifiedAt": "2026-10-18",
   "confidence": "official",
   "notes": "Snapshot gpt-5.4-mini-2026-03-17. Max input 272K. No long-context tier. Reasoning tokens billed as output."
  },
  {
   "id": "gemini-4-argon",
   "provider": "google",
   "name": "Gemini 4 Argon",
   "tier": "flagship",
   "input": 2,
   "output": 10,
   "cachedInput": 0.1,
   "cacheWrite": null,
   "batchDiscount": 0.5,
   "contextWindow": 1048576,
   "maxOutput": 65536,
   "longContext": null,
   "scheduledChange": {
    "effective": "2026-12-01",
    "input": 4,
    "output": 20,
    "cachedInput": 0.2
   },
   "released": "2026-10",
   "source": "https://ai.google.dev/gemini-api/docs/pricing",
   "verifiedAt": "2026-10-18",
   "confidence": "official",
   "notes": "Launch price through Nov 30, 2026; doubles to $4/$20 on Dec 1, 2026. Implicit caching only. Paid tier only. Max output 65,536 sync default, long decodes up to 1M. released = model page \"Latest update\" month. Output price includes thinking tokens. Explicit context caching also bills storage per 1M tokens/hour; Google has no separate cache-write price."
  },
  {
   "id": "gemini-3.8-flash",
   "provider": "google",
   "name": "Gemini 3.8 Flash",
   "tier": "mid",
   "input": 0.75,
   "output": 3.75,
   "cachedInput": 0.075,
   "cacheWrite": null,
   "batchDiscount": 0.5,
   "contextWindow": 1048576,
   "maxOutput": 65536,
   "longContext": null,
   "scheduledChange": {
    "effective": "2027-01-01",
    "input": 1.5,
    "output": 7.5,
    "cachedInput": 0.15
   },
   "released": "2026-09",
   "source": "https://ai.google.dev/gemini-api/docs/pricing",
   "verifiedAt": "2026-10-18",
   "confidence": "official",
   "notes": "Price through Dec 31, 2026; doubles Jan 1, 2027. Cache storage $0.50/1M tok/hr (rising to $1.00). released = model page \"Latest update\" month. Output price includes thinking tokens. Explicit context caching also bills storage per 1M tokens/hour; Google has no separate cache-write price."
  },
  {
   "id": "gemini-3.6-flash",
   "provider": "google",
   "name": "Gemini 3.6 Flash",
   "tier": "mid",
   "input": 0.75,
   "output": 3.75,
   "cachedInput": 0.075,
   "cacheWrite": null,
   "batchDiscount": 0.5,
   "contextWindow": 1048576,
   "maxOutput": 65536,
   "longContext": null,
   "scheduledChange": {
    "effective": "2027-01-01",
    "input": 1.5,
    "output": 7.5,
    "cachedInput": 0.15
   },
   "released": "2026-07",
   "source": "https://ai.google.dev/gemini-api/docs/pricing",
   "verifiedAt": "2026-10-18",
   "confidence": "official",
   "notes": "Previous-gen Flash. Same promo schedule as 3.8 Flash. released = model page \"Latest update\" month. Output price includes thinking tokens. Explicit context caching also bills storage per 1M tokens/hour; Google has no separate cache-write price."
  },
  {
   "id": "gemini-3.1-pro-preview",
   "provider": "google",
   "name": "Gemini 3.1 Pro (Preview)",
   "tier": "flagship",
   "input": 2,
   "output": 12,
   "cachedInput": 0.2,
   "cacheWrite": null,
   "batchDiscount": 0.5,
   "contextWindow": 1048576,
   "maxOutput": 65536,
   "longContext": {
    "threshold": 200000,
    "input": 4,
    "output": 18,
    "cachedInput": 0.4
   },
   "released": "2026-02",
   "source": "https://ai.google.dev/gemini-api/docs/pricing",
   "verifiedAt": "2026-10-18",
   "confidence": "official",
   "notes": "Preview. Tiered above 200K-token prompts. Cache storage $4.50/1M tok/hr. released = model page \"Latest update\" month. Output price includes thinking tokens. Explicit context caching also bills storage per 1M tokens/hour; Google has no separate cache-write price."
  },
  {
   "id": "gemini-3.5-flash-lite",
   "provider": "google",
   "name": "Gemini 3.5 Flash-Lite",
   "tier": "small",
   "input": 0.3,
   "output": 2.5,
   "cachedInput": 0.03,
   "cacheWrite": null,
   "batchDiscount": 0.5,
   "contextWindow": 1048576,
   "maxOutput": 65536,
   "longContext": null,
   "released": "2026-07",
   "source": "https://ai.google.dev/gemini-api/docs/pricing",
   "verifiedAt": "2026-10-18",
   "confidence": "official",
   "notes": "Same rate for text/image/video/audio input. Cache storage $1.00/1M tok/hr. Batch cache price $0.02. released = model page \"Latest update\" month. Output price includes thinking tokens. Explicit context caching also bills storage per 1M tokens/hour; Google has no separate cache-write price."
  },
  {
   "id": "gemini-3.1-flash-lite",
   "provider": "google",
   "name": "Gemini 3.1 Flash-Lite",
   "tier": "small",
   "input": 0.25,
   "output": 1.5,
   "cachedInput": 0.025,
   "cacheWrite": null,
   "batchDiscount": 0.5,
   "contextWindow": 1048576,
   "maxOutput": 65536,
   "longContext": null,
   "released": "2026-05",
   "source": "https://ai.google.dev/gemini-api/docs/pricing",
   "verifiedAt": "2026-10-18",
   "confidence": "official",
   "notes": "Text/image/video input price; audio input $0.50. released = model page \"Latest update\" month. Output price includes thinking tokens. Explicit context caching also bills storage per 1M tokens/hour; Google has no separate cache-write price."
  },
  {
   "id": "gemini-2.5-flash",
   "provider": "google",
   "name": "Gemini 2.5 Flash",
   "tier": "small",
   "input": 0.3,
   "output": 2.5,
   "cachedInput": 0.03,
   "cacheWrite": null,
   "batchDiscount": 0.5,
   "contextWindow": 1048576,
   "maxOutput": 65536,
   "longContext": null,
   "released": "2025-06",
   "source": "https://ai.google.dev/gemini-api/docs/pricing",
   "verifiedAt": "2026-10-18",
   "confidence": "official",
   "notes": "Legacy: Google limits access to projects that used it before; new projects should use 3.5 Flash-Lite / 3.8 Flash. Audio input $1.00. released = model page \"Latest update\" month. Output price includes thinking tokens. Explicit context caching also bills storage per 1M tokens/hour; Google has no separate cache-write price."
  },
  {
   "id": "grok-4.7",
   "provider": "xai",
   "name": "Grok 4.7",
   "tier": "flagship",
   "input": 2,
   "output": 6,
   "cachedInput": 0.5,
   "cacheWrite": null,
   "batchDiscount": null,
   "contextWindow": 500000,
   "maxOutput": null,
   "longContext": {
    "threshold": 200000,
    "input": 4,
    "output": 12,
    "cachedInput": 1
   },
   "released": null,
   "source": "https://docs.x.ai/developers/models/grok-4.7",
   "verifiedAt": "2026-10-18",
   "confidence": "official",
   "notes": "Reasoning effort low/medium/high/xhigh (default high). No Batch API support. Requests with >=200K prompt tokens bill ALL tokens at the higher rate. Grok 4.7 Fast is not on the public API."
  },
  {
   "id": "grok-4.3",
   "provider": "xai",
   "name": "Grok 4.3",
   "tier": "mid",
   "input": 1.25,
   "output": 2.5,
   "cachedInput": 0.2,
   "cacheWrite": null,
   "batchDiscount": 0.2,
   "contextWindow": 1000000,
   "maxOutput": null,
   "longContext": {
    "threshold": 200000,
    "input": 2.5,
    "output": 5,
    "cachedInput": 0.4
   },
   "released": null,
   "source": "https://docs.x.ai/developers/pricing",
   "verifiedAt": "2026-10-18",
   "confidence": "official",
   "notes": "Batch = 20% off (not 50%). Requests with >=200K prompt tokens bill ALL tokens at the higher rate. Grok 4.7 Fast is not on the public API."
  },
  {
   "id": "grok-4.20-0309-reasoning",
   "provider": "xai",
   "name": "Grok 4.20 (Reasoning)",
   "tier": "reasoning",
   "input": 1.25,
   "output": 2.5,
   "cachedInput": 0.2,
   "cacheWrite": null,
   "batchDiscount": 0.2,
   "contextWindow": 1000000,
   "maxOutput": null,
   "longContext": {
    "threshold": 200000,
    "input": 2.5,
    "output": 5,
    "cachedInput": 0.4
   },
   "released": null,
   "source": "https://docs.x.ai/developers/pricing",
   "verifiedAt": "2026-10-18",
   "confidence": "official",
   "notes": "Non-reasoning variant grok-4.20-0309-non-reasoning has identical prices. Batch = 20% off. Requests with >=200K prompt tokens bill ALL tokens at the higher rate. Grok 4.7 Fast is not on the public API."
  },
  {
   "id": "grok-build-0.1",
   "provider": "xai",
   "name": "Grok Build 0.1",
   "tier": "small",
   "input": 1,
   "output": 2,
   "cachedInput": 0.2,
   "cacheWrite": null,
   "batchDiscount": null,
   "contextWindow": 256000,
   "maxOutput": null,
   "longContext": {
    "threshold": 200000,
    "input": 2,
    "output": 4,
    "cachedInput": 0.4
   },
   "released": null,
   "source": "https://docs.x.ai/developers/pricing",
   "verifiedAt": "2026-10-18",
   "confidence": "official",
   "notes": "Coding model; xAI's cheapest public API text model. No batch discount. Requests with >=200K prompt tokens bill ALL tokens at the higher rate. Grok 4.7 Fast is not on the public API."
  },
  {
   "id": "deepseek-v4-pro",
   "provider": "deepseek",
   "name": "DeepSeek V4 Pro",
   "tier": "flagship",
   "input": 1.32,
   "output": 3.96,
   "cachedInput": 0.044,
   "cacheWrite": null,
   "batchDiscount": null,
   "contextWindow": 1000000,
   "maxOutput": 384000,
   "longContext": null,
   "offPeak": {
    "input": 0.66,
    "output": 1.98,
    "cachedInput": 0.022
   },
   "released": null,
   "source": "https://api-docs.deepseek.com/quick_start/pricing",
   "verifiedAt": "2026-10-18",
   "confidence": "official",
   "notes": "Version DeepSeek-V4-Pro-0813. Text only (no vision). Prices shown are PEAK (01:00-04:00 and 06:00-10:00 UTC Mon-Fri, excl. Chinese public holidays). All other hours (most of the week) are off-peak at 50% - see offPeak. No batch API."
  },
  {
   "id": "deepseek-flash",
   "provider": "deepseek",
   "name": "DeepSeek V4.1 Flash",
   "tier": "small",
   "input": 0.3,
   "output": 1.2,
   "cachedInput": 0.006,
   "cacheWrite": null,
   "batchDiscount": null,
   "contextWindow": 1000000,
   "maxOutput": 384000,
   "longContext": null,
   "offPeak": {
    "input": 0.15,
    "output": 0.6,
    "cachedInput": 0.003
   },
   "released": null,
   "source": "https://api-docs.deepseek.com/quick_start/pricing",
   "verifiedAt": "2026-10-18",
   "confidence": "official",
   "notes": "Version DeepSeek-V4.1-Flash-1017; supports vision. Legacy ids deepseek-v4-flash / deepseek-v4-flash-vision-exp route here at this price. Prices shown are PEAK (01:00-04:00 and 06:00-10:00 UTC Mon-Fri, excl. Chinese public holidays). All other hours (most of the week) are off-peak at 50% - see offPeak. No batch API."
  },
  {
   "id": "mistral-large-4",
   "provider": "mistral",
   "name": "Mistral Large 4",
   "tier": "flagship",
   "input": 0.68,
   "output": 2.09,
   "cachedInput": 0.07,
   "cacheWrite": null,
   "batchDiscount": 0.5,
   "contextWindow": 1000000,
   "maxOutput": null,
   "longContext": null,
   "regularPrice": {
    "input": 1.36,
    "output": 4.18,
    "cachedInput": 0.14
   },
   "released": "2026-10",
   "source": "https://docs.mistral.ai/models/mistral-large-4-0",
   "verifiedAt": "2026-10-18",
   "confidence": "official",
   "notes": "Public Preview, open weights (v26.10). Listed at a SALE price (50% off $1.36/$4.18); sale end date not published. Alias mistral-large-4-0. Batch 50% off per mistral.ai/pricing. Max output not published."
  },
  {
   "id": "mistral-medium-latest",
   "provider": "mistral",
   "name": "Mistral Medium 3.5",
   "tier": "mid",
   "input": 1.5,
   "output": 7.5,
   "cachedInput": 0.15,
   "cacheWrite": null,
   "batchDiscount": 0.5,
   "contextWindow": 256000,
   "maxOutput": null,
   "longContext": null,
   "released": "2026-04",
   "source": "https://docs.mistral.ai/models/mistral-medium-3-5-26-04",
   "verifiedAt": "2026-10-18",
   "confidence": "official",
   "notes": "Aliases mistral-medium-3-5, mistral-medium-3. Pricier than Large 4 sale price. Batch 50% off per mistral.ai/pricing. Max output not published."
  },
  {
   "id": "mistral-small-2603",
   "provider": "mistral",
   "name": "Mistral Small 4",
   "tier": "small",
   "input": 0.15,
   "output": 0.6,
   "cachedInput": 0.015,
   "cacheWrite": null,
   "batchDiscount": 0.5,
   "contextWindow": 256000,
   "maxOutput": null,
   "longContext": null,
   "released": "2026-03",
   "source": "https://docs.mistral.ai/models/mistral-small-4-0-26-03",
   "verifiedAt": "2026-10-18",
   "confidence": "official",
   "notes": "Alias mistral-small-latest. Apache 2.0. Batch 50% off per mistral.ai/pricing. Max output not published."
  },
  {
   "id": "ministral-8b-2512",
   "provider": "mistral",
   "name": "Ministral 3 8B",
   "tier": "small",
   "input": 0.15,
   "output": 0.15,
   "cachedInput": 0.015,
   "cacheWrite": null,
   "batchDiscount": 0.5,
   "contextWindow": 256000,
   "maxOutput": null,
   "longContext": null,
   "released": "2025-12",
   "source": "https://docs.mistral.ai/models/ministral-3-8b-25-12",
   "verifiedAt": "2026-10-18",
   "confidence": "official",
   "notes": "Alias ministral-8b-latest. 14B ($0.20/$0.20) and 3B ($0.10/$0.10) variants also offered. Batch 50% off per mistral.ai/pricing. Max output not published."
  },
  {
   "id": "codestral-2508",
   "provider": "mistral",
   "name": "Codestral",
   "tier": "small",
   "input": 0.3,
   "output": 0.9,
   "cachedInput": 0.03,
   "cacheWrite": null,
   "batchDiscount": 0.5,
   "contextWindow": 128000,
   "maxOutput": null,
   "longContext": null,
   "released": "2025-07",
   "source": "https://docs.mistral.ai/models/codestral-25-08",
   "verifiedAt": "2026-10-18",
   "confidence": "official",
   "notes": "Code model. Alias codestral-latest. Batch 50% off per mistral.ai/pricing. Max output not published."
  }
 ],
 "tools": {
  "anthropic": {
   "source": "https://platform.claude.com/docs/en/about-claude/pricing",
   "webSearch": {
    "per1kCalls": 10,
    "notes": "Plus token cost of search results (input tokens). Failed searches not billed."
   },
   "webFetch": {
    "per1kCalls": 0,
    "notes": "No per-call fee; fetched content billed as input tokens."
   },
   "codeExecution": {
    "perContainerHour": 0.05,
    "freeHoursPerMonth": 1550,
    "minimumMinutes": 5,
    "notes": "Free when web_search_20260209+ or web_fetch_20260209+ is in the request."
   }
  },
  "openai": {
   "source": "https://developers.openai.com/api/docs/pricing",
   "webSearch": {
    "per1kCalls": 10,
    "notes": "Plus search content tokens at model rates. web_search_preview on non-reasoning models: $25/1K, content tokens free."
   },
   "fileSearch": {
    "per1kCalls": 2.5,
    "storagePerGBDay": 0.1,
    "freeStorageGB": 1
   },
   "codeInterpreter": {
    "per20MinSession": {
     "1GB": 0.03,
     "4GB": 0.12,
     "16GB": 0.48,
     "64GB": 1.92
    },
    "notes": "Hosted Shell / Code Interpreter containers; billed per minute, 5-min minimum."
   }
  },
  "google": {
   "source": "https://ai.google.dev/gemini-api/docs/pricing",
   "groundingWithGoogleSearch": {
    "per1kCalls": 14,
    "freePerMonth": 5000,
    "notes": "Gemini 3 and newer; one prompt may trigger several billed search queries. Gemini 2.5: $35/1K grounded prompts after 1,500 free/day."
   },
   "groundingWithGoogleMaps": {
    "per1kCalls": 14,
    "freePerMonth": 5000
   }
  },
  "xai": {
   "source": "https://docs.x.ai/developers/pricing",
   "webSearch": {
    "per1kCalls": 5
   },
   "xSearch": {
    "per1kPosts": 5,
    "per1kProfiles": 10
   },
   "codeExecution": {
    "per1kCalls": 5
   },
   "fileAttachmentSearch": {
    "per1kCalls": 5
   },
   "collectionsSearch": {
    "per1kCalls": 2.5
   }
  },
  "deepseek": {
   "source": "https://api-docs.deepseek.com/quick_start/pricing",
   "search": {
    "per1kCalls": 0.75,
    "notes": "Per successful request; billing starts 2026-11-01 00:00 Beijing time."
   },
   "fetch": {
    "per1kCalls": 0.75,
    "notes": "Same as search."
   }
  }
 }
}