{
  "category": "model_providers",
  "cbt_reference": {
    "assumption": "75% input / 25% output",
    "ct_value_usd": 0.0001
  },
  "items": [
    {
      "blended_usd_per_1m": 0.1312,
      "cbt_blended_per_1m": 1312,
      "cbt_input_per_1m": 750,
      "cbt_output_per_1m": 3000,
      "cost_input_per_1k_usd": 7.5e-05,
      "cost_input_usd_per_1m": 0.075,
      "cost_output_per_1k_usd": 0.0003,
      "cost_output_usd_per_1m": 0.3,
      "id": "vertex-gemini-2.5-flash-lite",
      "last_verified": "2026-03-08",
      "max_output_tokens": 8192,
      "model_label": "Gemini 2.5 Flash-Lite",
      "notes": "Current runtime target in the Coolbits model registry.",
      "provider_model_id": "gemini-2.5-flash-lite",
      "provider_name": "Google Vertex AI",
      "provider_slug": "vertex",
      "role": "Fast chat and routing",
      "status": "live",
      "status_label": "Live & Verified"
    },
    {
      "blended_usd_per_1m": 4.375,
      "cbt_blended_per_1m": 43750,
      "cbt_input_per_1m": 35000,
      "cbt_output_per_1m": 70000,
      "cost_input_per_1k_usd": 0.0035,
      "cost_input_usd_per_1m": 3.5,
      "cost_output_per_1k_usd": 0.007,
      "cost_output_usd_per_1m": 7.0,
      "id": "vertex-gemini-2.5-pro",
      "last_verified": "2026-03-08",
      "max_output_tokens": 8192,
      "model_label": "Gemini 2.5 Pro",
      "notes": "Current runtime target in the Coolbits model registry.",
      "provider_model_id": "gemini-2.5-pro",
      "provider_name": "Google Vertex AI",
      "provider_slug": "vertex",
      "role": "Deep analysis",
      "status": "live",
      "status_label": "Live & Verified"
    },
    {
      "blended_usd_per_1m": 0.2625,
      "cbt_blended_per_1m": 2625,
      "cbt_input_per_1m": 1500,
      "cbt_output_per_1m": 6000,
      "cost_input_per_1k_usd": 0.00015,
      "cost_input_usd_per_1m": 0.15,
      "cost_output_per_1k_usd": 0.0006,
      "cost_output_usd_per_1m": 0.6,
      "id": "openai-gpt-4.1-mini",
      "last_verified": "2026-03-08",
      "max_output_tokens": 16384,
      "model_label": "GPT-4.1 mini",
      "notes": "Server-side only; shown as the current provider target, not a browser key flow.",
      "provider_model_id": "gpt-4.1-mini",
      "provider_name": "OpenAI",
      "provider_slug": "openai",
      "role": "Light reasoning and routing",
      "status": "live",
      "status_label": "Live & Verified"
    },
    {
      "blended_usd_per_1m": 7.5,
      "cbt_blended_per_1m": 75000,
      "cbt_input_per_1m": 50000,
      "cbt_output_per_1m": 150000,
      "cost_input_per_1k_usd": 0.005,
      "cost_input_usd_per_1m": 5.0,
      "cost_output_per_1k_usd": 0.015,
      "cost_output_usd_per_1m": 15.0,
      "id": "openai-gpt-4.1",
      "last_verified": "2026-03-08",
      "max_output_tokens": 8192,
      "model_label": "GPT-4.1",
      "notes": "Server-side only; managed or BYOK billing is resolved inside the backend layer.",
      "provider_model_id": "gpt-4.1",
      "provider_name": "OpenAI",
      "provider_slug": "openai",
      "role": "General reasoning",
      "status": "live",
      "status_label": "Live & Verified"
    },
    {
      "blended_usd_per_1m": 6.0,
      "cbt_blended_per_1m": 60000,
      "cbt_input_per_1m": 30000,
      "cbt_output_per_1m": 150000,
      "cost_input_per_1k_usd": 0.003,
      "cost_input_usd_per_1m": 3.0,
      "cost_output_per_1k_usd": 0.015,
      "cost_output_usd_per_1m": 15.0,
      "id": "anthropic-claude-sonnet",
      "last_verified": "2026-03-08",
      "max_output_tokens": 8192,
      "model_label": "Claude Sonnet 4",
      "notes": "Current runtime target in the Coolbits model registry.",
      "provider_model_id": "claude-sonnet-4-20250514",
      "provider_name": "Anthropic",
      "provider_slug": "anthropic",
      "role": "Balanced reasoning and writing",
      "status": "live",
      "status_label": "Live & Verified"
    },
    {
      "blended_usd_per_1m": 1.6,
      "cbt_blended_per_1m": 16000,
      "cbt_input_per_1m": 8000,
      "cbt_output_per_1m": 40000,
      "cost_input_per_1k_usd": 0.0008,
      "cost_input_usd_per_1m": 0.8,
      "cost_output_per_1k_usd": 0.004,
      "cost_output_usd_per_1m": 4.0,
      "id": "anthropic-claude-haiku",
      "last_verified": "2026-03-08",
      "max_output_tokens": 8192,
      "model_label": "Claude 3.5 Haiku",
      "notes": "Current runtime target in the Coolbits model registry.",
      "provider_model_id": "claude-3-5-haiku-20241022",
      "provider_name": "Anthropic",
      "provider_slug": "anthropic",
      "role": "Fast tasks",
      "status": "live",
      "status_label": "Live & Verified"
    },
    {
      "blended_usd_per_1m": 0.35,
      "cbt_blended_per_1m": 3500,
      "cbt_input_per_1m": 3000,
      "cbt_output_per_1m": 5000,
      "cost_input_per_1k_usd": 0.0003,
      "cost_input_usd_per_1m": 0.3,
      "cost_output_per_1k_usd": 0.0005,
      "cost_output_usd_per_1m": 0.5,
      "id": "xai-grok-3-mini",
      "last_verified": "2026-03-08",
      "max_output_tokens": 8192,
      "model_label": "Grok 3 Mini Fast",
      "notes": "Current runtime target in the Coolbits model registry.",
      "provider_model_id": "grok-3-mini-fast",
      "provider_name": "xAI",
      "provider_slug": "xai",
      "role": "Fast reasoning",
      "status": "live",
      "status_label": "Live & Verified"
    },
    {
      "blended_usd_per_1m": 6.0,
      "cbt_blended_per_1m": 60000,
      "cbt_input_per_1m": 30000,
      "cbt_output_per_1m": 150000,
      "cost_input_per_1k_usd": 0.003,
      "cost_input_usd_per_1m": 3.0,
      "cost_output_per_1k_usd": 0.015,
      "cost_output_usd_per_1m": 15.0,
      "id": "xai-grok-3",
      "last_verified": "2026-03-08",
      "max_output_tokens": 16384,
      "model_label": "Grok 3",
      "notes": "Current runtime target in the Coolbits model registry.",
      "provider_model_id": "grok-3",
      "provider_name": "xAI",
      "provider_slug": "xai",
      "role": "Deep analysis",
      "status": "live",
      "status_label": "Live & Verified"
    }
  ],
  "last_updated": "2026-03-08",
  "ok": true
}