{
  "$schema": "https://json-schema.org/draft/2020-12/schema",
  "dataset": "LLM API list prices, corrected for prompt caching and batch tiers",
  "publisher": "rabayid.com",
  "homepage": "https://rabayid.com/",
  "human_page": "https://rabayid.com/",
  "machine_page": "https://rabayid.com/data/pricing.json",
  "prices_verified": "September 2026",
  "generated": "2026-09-26",
  "currency": "USD",
  "unit": "per 1,000,000 tokens",
  "model_count": 25,
  "providers": [
    "Anthropic",
    "DeepSeek",
    "Google",
    "OpenAI",
    "xAI"
  ],
  "field_notes": {
    "input": "Price for input tokens that were not served from cache.",
    "cached_input": "Price for input tokens read from a warm prompt cache. Cheapest billed input class.",
    "cache_write": "Price for writing tokens into cache. Carries a premium on Anthropic. Not published by Google, which bills cached-context storage per MTok-hour instead.",
    "output": "Price for generated tokens, including reasoning tokens where the provider bills them as output.",
    "supports_batch_api_false": "Provider publishes no batch tier. Such models are excluded from batch rankings rather than shown at zero cost.",
    "long_context_tier": "Providers switch to these rates once input exceeds the threshold. Quoting a single flat rate is misleading for large prompts."
  },
  "accuracy_note": "Transcribed from each provider's own published pricing page. Rates change often and several are promotional. Always confirm against the provider before committing to a contract.",
  "models": [
    {
      "id": "gpt-6-astra",
      "name": "GPT-6 Astra",
      "provider": "OpenAI",
      "context_window_tokens": 1100000,
      "usd_per_million_tokens": {
        "input": 10,
        "cached_input": 1,
        "cache_write": 12.5,
        "output": 50
      },
      "supports_batch_api": true,
      "promotional": false,
      "flagship": true,
      "long_context_tier": {
        "threshold_tokens": 272000,
        "usd_per_million_tokens": {
          "input": 20,
          "cached_input": 2,
          "cache_write": 25,
          "output": 75
        }
      },
      "note": "Flagship. Released Sep 2026."
    },
    {
      "id": "gpt-5.6-sol",
      "name": "GPT-5.6 Sol",
      "provider": "OpenAI",
      "context_window_tokens": 1100000,
      "usd_per_million_tokens": {
        "input": 4,
        "cached_input": 0.4,
        "cache_write": 5,
        "output": 20
      },
      "supports_batch_api": true,
      "promotional": true,
      "promotion_ends": "Nov 21, 2026",
      "flagship": true,
      "long_context_tier": {
        "threshold_tokens": 272000,
        "usd_per_million_tokens": {
          "input": 8,
          "cached_input": 0.8,
          "cache_write": 10,
          "output": 30
        }
      },
      "note": "Promotional rate, holds at least through Nov 21, 2026 (was $5/$30)."
    },
    {
      "id": "gpt-5.6-terra",
      "name": "GPT-5.6 Terra",
      "provider": "OpenAI",
      "context_window_tokens": 1100000,
      "usd_per_million_tokens": {
        "input": 2,
        "cached_input": 0.2,
        "cache_write": 2.5,
        "output": 12
      },
      "supports_batch_api": true,
      "promotional": false,
      "flagship": true,
      "long_context_tier": {
        "threshold_tokens": 272000,
        "usd_per_million_tokens": {
          "input": 4,
          "cached_input": 0.4,
          "cache_write": 5,
          "output": 18
        }
      },
      "note": "Default workhorse. Price cut Jul 30, 2026."
    },
    {
      "id": "gpt-5.6-luna",
      "name": "GPT-5.6 Luna",
      "provider": "OpenAI",
      "context_window_tokens": 1100000,
      "usd_per_million_tokens": {
        "input": 0.2,
        "cached_input": 0.02,
        "cache_write": 0.25,
        "output": 1.2
      },
      "supports_batch_api": true,
      "promotional": false,
      "long_context_tier": {
        "threshold_tokens": 272000,
        "usd_per_million_tokens": {
          "input": 0.4,
          "cached_input": 0.04,
          "cache_write": 0.5,
          "output": 1.8
        }
      },
      "note": "Budget tier. Classification, routing, extraction."
    },
    {
      "id": "gpt-5.5",
      "name": "GPT-5.5",
      "provider": "OpenAI",
      "context_window_tokens": 1100000,
      "usd_per_million_tokens": {
        "input": 5,
        "cached_input": 0.5,
        "cache_write": null,
        "output": 30
      },
      "supports_batch_api": true,
      "promotional": false,
      "flagship": true,
      "long_context_tier": {
        "threshold_tokens": 272000,
        "usd_per_million_tokens": {
          "input": 10,
          "cached_input": 1,
          "cache_write": null,
          "output": 45
        }
      },
      "note": "Previous flagship."
    },
    {
      "id": "gpt-5.4",
      "name": "GPT-5.4",
      "provider": "OpenAI",
      "context_window_tokens": 1100000,
      "usd_per_million_tokens": {
        "input": 2.5,
        "cached_input": 0.25,
        "cache_write": null,
        "output": 15
      },
      "supports_batch_api": true,
      "promotional": false,
      "long_context_tier": {
        "threshold_tokens": 272000,
        "usd_per_million_tokens": {
          "input": 5,
          "cached_input": 0.5,
          "cache_write": null,
          "output": 22.5
        }
      },
      "note": "Superseded by Terra."
    },
    {
      "id": "gpt-5.4-mini",
      "name": "GPT-5.4 mini",
      "provider": "OpenAI",
      "context_window_tokens": 400000,
      "usd_per_million_tokens": {
        "input": 0.75,
        "cached_input": 0.075,
        "cache_write": null,
        "output": 4.5
      },
      "supports_batch_api": true,
      "promotional": false,
      "note": "Lightweight tier."
    },
    {
      "id": "gpt-5.4-nano",
      "name": "GPT-5.4 nano",
      "provider": "OpenAI",
      "context_window_tokens": 400000,
      "usd_per_million_tokens": {
        "input": 0.2,
        "cached_input": 0.02,
        "cache_write": null,
        "output": 1.25
      },
      "supports_batch_api": true,
      "promotional": false,
      "note": "Cheapest GPT tier."
    },
    {
      "id": "gpt-5.2",
      "name": "GPT-5.2",
      "provider": "OpenAI",
      "context_window_tokens": 400000,
      "usd_per_million_tokens": {
        "input": 1.75,
        "cached_input": 0.175,
        "cache_write": null,
        "output": 14
      },
      "supports_batch_api": true,
      "promotional": false
    },
    {
      "id": "gpt-4.1",
      "name": "GPT-4.1",
      "provider": "OpenAI",
      "context_window_tokens": 400000,
      "usd_per_million_tokens": {
        "input": 2,
        "cached_input": 0.5,
        "cache_write": null,
        "output": 8
      },
      "supports_batch_api": true,
      "promotional": false,
      "note": "Still widely deployed."
    },
    {
      "id": "claude-fable-5.1",
      "name": "Claude Fable 5.1",
      "provider": "Anthropic",
      "context_window_tokens": 1000000,
      "usd_per_million_tokens": {
        "input": 10,
        "cached_input": 0.25,
        "cache_write": 12.5,
        "output": 50
      },
      "supports_batch_api": true,
      "promotional": false,
      "flagship": true,
      "note": "Released Sep 1, 2026. Cache read cut 75% ($1.00 → $0.25)."
    },
    {
      "id": "claude-opus-5",
      "name": "Claude Opus 5",
      "provider": "Anthropic",
      "context_window_tokens": 1000000,
      "usd_per_million_tokens": {
        "input": 5,
        "cached_input": 0.5,
        "cache_write": 6.25,
        "output": 25
      },
      "supports_batch_api": true,
      "promotional": false,
      "flagship": true,
      "note": "Released Jul 24, 2026. Same rate as Opus 4.8."
    },
    {
      "id": "claude-sonnet-5",
      "name": "Claude Sonnet 5",
      "provider": "Anthropic",
      "context_window_tokens": 1000000,
      "usd_per_million_tokens": {
        "input": 2,
        "cached_input": 0.2,
        "cache_write": 2.5,
        "output": 10
      },
      "supports_batch_api": true,
      "promotional": false,
      "flagship": true,
      "note": "Workhorse. $2/$10 confirmed permanent Aug 10, 2026."
    },
    {
      "id": "claude-sonnet-4.6",
      "name": "Claude Sonnet 4.6",
      "provider": "Anthropic",
      "context_window_tokens": 1000000,
      "usd_per_million_tokens": {
        "input": 3,
        "cached_input": 0.3,
        "cache_write": 3.75,
        "output": 15
      },
      "supports_batch_api": true,
      "promotional": false,
      "note": "Older balanced model."
    },
    {
      "id": "claude-haiku-4.5",
      "name": "Claude Haiku 4.5",
      "provider": "Anthropic",
      "context_window_tokens": 200000,
      "usd_per_million_tokens": {
        "input": 1,
        "cached_input": 0.1,
        "cache_write": 1.25,
        "output": 5
      },
      "supports_batch_api": true,
      "promotional": false,
      "note": "High-volume, low latency."
    },
    {
      "id": "gemini-3.8-flash",
      "name": "Gemini 3.8 Flash",
      "provider": "Google",
      "context_window_tokens": 1000000,
      "usd_per_million_tokens": {
        "input": 0.75,
        "cached_input": 0.075,
        "cache_write": null,
        "output": 3.75
      },
      "supports_batch_api": true,
      "promotional": true,
      "promotion_ends": "Dec 31, 2026",
      "long_context_tier": {
        "threshold_tokens": 200000,
        "usd_per_million_tokens": {
          "input": 1.5,
          "cached_input": null,
          "cache_write": null,
          "output": 7.5
        }
      },
      "note": "Released Sep 2, 2026. Introductory price, doubles Jan 1, 2027."
    },
    {
      "id": "gemini-3.7-flash",
      "name": "Gemini 3.7 Flash",
      "provider": "Google",
      "context_window_tokens": 1000000,
      "usd_per_million_tokens": {
        "input": 0.75,
        "cached_input": 0.075,
        "cache_write": null,
        "output": 3.75
      },
      "supports_batch_api": true,
      "promotional": true,
      "promotion_ends": "Dec 31, 2026",
      "long_context_tier": {
        "threshold_tokens": 200000,
        "usd_per_million_tokens": {
          "input": 1.5,
          "cached_input": null,
          "cache_write": null,
          "output": 7.5
        }
      },
      "note": "Most capable Flash for coding and agents."
    },
    {
      "id": "gemini-3.5-flash-lite",
      "name": "Gemini 3.5 Flash-Lite",
      "provider": "Google",
      "context_window_tokens": 1000000,
      "usd_per_million_tokens": {
        "input": 0.3,
        "cached_input": 0.03,
        "cache_write": null,
        "output": 2.5
      },
      "supports_batch_api": true,
      "promotional": false,
      "note": "Cheapest Gemini 3.x."
    },
    {
      "id": "gemini-3.5-flash",
      "name": "Gemini 3.5 Flash",
      "provider": "Google",
      "context_window_tokens": 1000000,
      "usd_per_million_tokens": {
        "input": 1.5,
        "cached_input": 0.15,
        "cache_write": null,
        "output": 9
      },
      "supports_batch_api": true,
      "promotional": false,
      "note": "GA May 2026."
    },
    {
      "id": "gemini-3.1-pro",
      "name": "Gemini 3.1 Pro (Preview)",
      "provider": "Google",
      "context_window_tokens": 1000000,
      "usd_per_million_tokens": {
        "input": 2,
        "cached_input": 0.2,
        "cache_write": null,
        "output": 12
      },
      "supports_batch_api": true,
      "promotional": false,
      "long_context_tier": {
        "threshold_tokens": 200000,
        "usd_per_million_tokens": {
          "input": 4,
          "cached_input": null,
          "cache_write": null,
          "output": 18
        }
      },
      "note": "Preview pricing. Tiers above 200K context."
    },
    {
      "id": "gemini-2.5-pro",
      "name": "Gemini 2.5 Pro",
      "provider": "Google",
      "context_window_tokens": 1000000,
      "usd_per_million_tokens": {
        "input": 1.25,
        "cached_input": 0.125,
        "cache_write": null,
        "output": 10
      },
      "supports_batch_api": true,
      "promotional": false,
      "long_context_tier": {
        "threshold_tokens": 200000,
        "usd_per_million_tokens": {
          "input": 2.5,
          "cached_input": null,
          "cache_write": null,
          "output": 15
        }
      },
      "note": "Tiers above 200K context."
    },
    {
      "id": "grok-4.6",
      "name": "Grok 4.6",
      "provider": "xAI",
      "context_window_tokens": 524000,
      "usd_per_million_tokens": {
        "input": 2,
        "cached_input": 0.2,
        "cache_write": null,
        "output": 6
      },
      "supports_batch_api": true,
      "promotional": false,
      "long_context_tier": {
        "threshold_tokens": 200000,
        "usd_per_million_tokens": {
          "input": 4,
          "cached_input": null,
          "cache_write": null,
          "output": 12
        }
      },
      "note": "Tiers above 200K context."
    },
    {
      "id": "grok-4.3",
      "name": "Grok 4.3",
      "provider": "xAI",
      "context_window_tokens": 1000000,
      "usd_per_million_tokens": {
        "input": 1.25,
        "cached_input": 0.125,
        "cache_write": null,
        "output": 2.5
      },
      "supports_batch_api": true,
      "promotional": false,
      "long_context_tier": {
        "threshold_tokens": 200000,
        "usd_per_million_tokens": {
          "input": 2.5,
          "cached_input": null,
          "cache_write": null,
          "output": 5
        }
      },
      "note": "Tiers above 200K context."
    },
    {
      "id": "deepseek-v4.1-flash",
      "name": "DeepSeek V4.1 Flash",
      "provider": "DeepSeek",
      "context_window_tokens": 1100000,
      "usd_per_million_tokens": {
        "input": 0.13,
        "cached_input": 0.0028,
        "cache_write": null,
        "output": 0.28
      },
      "supports_batch_api": false,
      "promotional": false,
      "note": "Open weights. Cache read ~2% of list. No batch tier."
    },
    {
      "id": "deepseek-v4-pro",
      "name": "DeepSeek V4 Pro",
      "provider": "DeepSeek",
      "context_window_tokens": 1000000,
      "usd_per_million_tokens": {
        "input": 0.435,
        "cached_input": 0.0036,
        "cache_write": null,
        "output": 0.87
      },
      "supports_batch_api": false,
      "promotional": false,
      "note": "Open weights. No batch tier."
    }
  ]
}