{
  "metadata": {
    "version": "2.3.0",
    "name": "LLM Cost Engine Pricing Registry",
    "last_updated": "2026-02-24",
    "last_verified": "2026-02-24",
    "base_currency": "USD",
    "pricing_unit": "per_1M_tokens",
    "maintained_by": "LLM Cost Engine Team",
    "source_verification": "Official provider pricing pages",
    "update_frequency": "weekly"
  },
  "preset_profiles": {
    "caching_default": { "tier": "standard", "provider": "Anthropic" },
    "batch_default":   { "tier": "standard", "provider": "OpenAI" },
    "chatbot_defaults": [
      { "tier": "flagship",  "provider": "OpenAI" },
      { "tier": "standard",  "provider": "Anthropic" },
      { "tier": "efficient", "provider": "Google" },
      { "tier": "standard",  "provider": "DeepSeek" },
      { "tier": "standard",  "provider": "Meta" }
    ]
  },
  "models": [
    {
      "id": "gpt-5.2",
      "name": "GPT-5.2",
      "provider": "OpenAI",
      "tier": "flagship",
      "provider_rank": 1,
      "tags": ["chat", "code", "reasoning"],
      "pricing": {
        "input_1m": 1.75,
        "output_1m": 14.0,
        "cached_input_1m": 0.175,
        "batch_input_1m": 0.875,
        "batch_output_1m": 7.0
      },
      "capabilities": {
        "context_window": 128000,
        "latency_index": 0.93
      }
    },
    {
      "id": "gpt-5.1",
      "name": "GPT-5.1",
      "provider": "OpenAI",
      "tier": "flagship",
      "provider_rank": 2,
      "tags": ["chat", "code", "reasoning"],
      "pricing": {
        "input_1m": 1.25,
        "output_1m": 10.0,
        "cached_input_1m": 0.125,
        "batch_input_1m": 0.625,
        "batch_output_1m": 5.0
      },
      "capabilities": {
        "context_window": 128000,
        "latency_index": 0.9
      }
    },
    {
      "id": "gpt-5-mini",
      "name": "GPT-5 Mini",
      "provider": "OpenAI",
      "tier": "mini",
      "provider_rank": 3,
      "tags": ["chat", "content"],
      "pricing": {
        "input_1m": 0.25,
        "output_1m": 2.0,
        "cached_input_1m": 0.025,
        "batch_input_1m": 0.125,
        "batch_output_1m": 1.0
      },
      "capabilities": {
        "context_window": 128000,
        "latency_index": 0.96
      }
    },
    {
      "id": "o3-mini",
      "name": "o3-mini",
      "provider": "OpenAI",
      "tier": "efficient",
      "provider_rank": 4,
      "tags": ["reasoning", "code"],
      "pricing": {
        "input_1m": 1.1,
        "output_1m": 4.4,
        "cached_input_1m": 0.55,
        "batch_input_1m": 0.55,
        "batch_output_1m": 2.2
      },
      "capabilities": {
        "context_window": 200000,
        "latency_index": 0.8
      }
    },
    {
      "id": "claude-opus-4.6",
      "name": "Claude Opus 4.6",
      "provider": "Anthropic",
      "tier": "flagship",
      "provider_rank": 1,
      "tags": ["chat", "code", "reasoning", "rag"],
      "pricing": {
        "input_1m": 5.0,
        "output_1m": 25.0,
        "cached_input_1m": 0.5,
        "batch_input_1m": 2.5,
        "batch_output_1m": 12.5
      },
      "capabilities": {
        "context_window": 200000,
        "latency_index": 0.85
      }
    },
    {
      "id": "claude-sonnet-4.6",
      "name": "Claude Sonnet 4.6",
      "provider": "Anthropic",
      "tier": "standard",
      "provider_rank": 2,
      "tags": ["chat", "code", "reasoning", "rag"],
      "pricing": {
        "input_1m": 3.0,
        "output_1m": 15.0,
        "cached_input_1m": 0.3,
        "batch_input_1m": 1.5,
        "batch_output_1m": 7.5
      },
      "capabilities": {
        "context_window": 200000,
        "latency_index": 0.95
      }
    },
    {
      "id": "claude-haiku-4.5",
      "name": "Claude Haiku 4.5",
      "provider": "Anthropic",
      "tier": "mini",
      "provider_rank": 3,
      "tags": ["chat", "content"],
      "pricing": {
        "input_1m": 1.0,
        "output_1m": 5.0,
        "cached_input_1m": 0.1,
        "batch_input_1m": 0.5,
        "batch_output_1m": 2.5
      },
      "capabilities": {
        "context_window": 200000,
        "latency_index": 0.99
      }
    },
    {
      "id": "gemini-3.1-pro",
      "name": "Gemini 3.1 Pro",
      "provider": "Google",
      "tier": "flagship",
      "provider_rank": 1,
      "tags": ["chat", "long-context", "reasoning"],
      "pricing": {
        "input_1m": 2.0,
        "output_1m": 12.0,
        "cached_input_1m": 0.2
      },
      "capabilities": {
        "context_window": 1000000,
        "latency_index": 0.92
      }
    },
    {
      "id": "gemini-3-flash",
      "name": "Gemini 3 Flash",
      "provider": "Google",
      "tier": "efficient",
      "provider_rank": 2,
      "tags": ["chat", "long-context", "rag"],
      "pricing": {
        "input_1m": 0.5,
        "output_1m": 3.0,
        "cached_input_1m": 0.125
      },
      "capabilities": {
        "context_window": 1000000,
        "latency_index": 0.98
      }
    },
    {
      "id": "deepseek-r1",
      "name": "DeepSeek R1",
      "provider": "DeepSeek",
      "tier": "flagship",
      "provider_rank": 1,
      "tags": ["reasoning", "code"],
      "pricing": {
        "input_1m": 0.55,
        "output_1m": 2.19,
        "cached_input_1m": 0.14
      },
      "capabilities": {
        "context_window": 128000,
        "latency_index": 0.75
      }
    },
    {
      "id": "deepseek-v3",
      "name": "DeepSeek V3",
      "provider": "DeepSeek",
      "tier": "standard",
      "provider_rank": 2,
      "tags": ["chat", "code", "content"],
      "pricing": {
        "input_1m": 0.28,
        "output_1m": 0.42,
        "cached_input_1m": 0.028
      },
      "capabilities": {
        "context_window": 128000,
        "latency_index": 0.88
      }
    },
    {
      "id": "llama-4-maverick",
      "name": "Llama 4 Maverick",
      "provider": "Meta",
      "tier": "standard",
      "provider_rank": 1,
      "tags": ["chat", "content"],
      "pricing": {
        "input_1m": 0.20,
        "output_1m": 0.60
      },
      "capabilities": {
        "context_window": 128000,
        "latency_index": 0.95
      }
    },
    {
      "id": "llama-3.3-70b",
      "name": "Llama 3.3 70B",
      "provider": "Meta",
      "tier": "standard",
      "provider_rank": 2,
      "tags": ["chat", "code"],
      "pricing": {
        "input_1m": 0.59,
        "output_1m": 0.79
      },
      "capabilities": {
        "context_window": 128000,
        "latency_index": 0.88
      }
    },
    {
      "id": "mistral-large",
      "name": "Mistral Large 3",
      "provider": "Mistral AI",
      "tier": "flagship",
      "provider_rank": 1,
      "tags": ["chat", "code", "reasoning"],
      "pricing": {
        "input_1m": 0.5,
        "output_1m": 1.5,
        "cached_input_1m": 0.25
      },
      "capabilities": {
        "context_window": 128000,
        "latency_index": 0.85
      }
    },
    {
      "id": "mistral-medium-3",
      "name": "Mistral Medium 3.1",
      "provider": "Mistral AI",
      "tier": "standard",
      "provider_rank": 2,
      "tags": ["chat", "content"],
      "pricing": {
        "input_1m": 0.4,
        "output_1m": 2.0
      },
      "capabilities": {
        "context_window": 128000,
        "latency_index": 0.88
      }
    },
    {
      "id": "mistral-small",
      "name": "Mistral Small 3.2",
      "provider": "Mistral AI",
      "tier": "efficient",
      "provider_rank": 3,
      "tags": ["chat", "content"],
      "pricing": {
        "input_1m": 0.1,
        "output_1m": 0.3,
        "cached_input_1m": 0.05
      },
      "capabilities": {
        "context_window": 128000,
        "latency_index": 0.92
      }
    },
    {
      "id": "qwen3.5-plus",
      "name": "Qwen 3.5-Plus",
      "provider": "Qwen",
      "tier": "flagship",
      "provider_rank": 1,
      "tags": ["chat", "code", "reasoning", "long-context"],
      "pricing": {
        "input_1m": 0.40,
        "output_1m": 2.40
      },
      "capabilities": {
        "context_window": 262144,
        "latency_index": 0.87
      }
    }
  ]
}