{
  "_meta": {
    "source": "https://apipriceindex.com/api/deepinfra/qwen-2.5-72b",
    "attribution": "apipriceindex.com",
    "license": "CC BY 4.0",
    "license_url": "https://creativecommons.org/licenses/by/4.0/",
    "note": "Free to reuse with attribution to apipriceindex.com.",
    "generated_at": "2026-09-05T08:15:05.874Z"
  },
  "id": "deepinfra:qwen-2.5-72b",
  "name": "Qwen 2.5 72B (DeepInfra)",
  "base_model": "qwen-2.5-72b",
  "provider": {
    "id": "deepinfra",
    "name": "DeepInfra",
    "url": "https://deepinfra.com"
  },
  "category": "llm-chat",
  "capabilities": [
    "text",
    "function-calling"
  ],
  "context_window": 131072,
  "max_output": 32768,
  "pricing": {
    "model_id": "deepinfra:qwen-2.5-72b",
    "input": 0.36,
    "output": 0.4,
    "cached_input": null,
    "source_url": "https://openrouter.ai/qwen/qwen-2.5-72b-instruct",
    "verified_at": "2026-09-05T08:00:57Z",
    "sources": [
      {
        "name": "OpenRouter",
        "input": 0.36,
        "output": 0.4,
        "url": "https://openrouter.ai/qwen/qwen-2.5-72b-instruct",
        "verified_at": "2026-09-05T08:00:57Z"
      },
      {
        "name": "LiteLLM",
        "input": 0.38,
        "output": 0.4,
        "url": "https://github.com/BerriAI/litellm/blob/main/model_prices_and_context_window.json",
        "verified_at": "2026-09-03T20:47:46Z"
      },
      {
        "name": "models.dev",
        "input": 0.36,
        "output": 0.4,
        "url": "https://models.dev/",
        "verified_at": "2026-09-04T10:52:49Z"
      }
    ],
    "confidence": "corroborated",
    "delta_pct": 0
  },
  "price_history": [
    [
      "2026-09-03",
      0.36,
      0.4
    ]
  ],
  "latency": [],
  "license": {
    "license": "Apache 2.0",
    "tier": "permissive",
    "open_weights": true,
    "source_url": "https://huggingface.co/Qwen/Qwen3-32B/blob/main/LICENSE"
  },
  "cutoff": {
    "cutoff": "2025-09",
    "label": "September 2025",
    "source_url": "https://aiknowledgecutoff.com/"
  },
  "intelligence": null,
  "quality": {
    "elo": 1303,
    "arena_name": "Qwen2.5-72B",
    "verified_at": "2026-09-04",
    "benchmark_label": "LMArena (Chatbot Arena)",
    "source_url": "https://lmarena.ai/leaderboard/text"
  },
  "api": {
    "base_url": "https://openrouter.ai/api/v1",
    "model": "qwen/qwen-2.5-72b-instruct",
    "via": "openrouter",
    "key_env": "OPENROUTER_API_KEY",
    "docs_url": "https://openrouter.ai/qwen/qwen-2.5-72b-instruct",
    "or_provider": "DeepInfra"
  },
  "facts": [
    "Qwen 2.5 72B (DeepInfra) costs $0.36 input / $0.40 output per 1M tokens.",
    "A chatbot workload (1000 in / 500 out tokens) costs $0.56 per 1,000 requests.",
    "Cross-checked against LiteLLM's public price index, which lists qwen-2.5-72b at $0.38/$0.40 per 1M (within 3% of this endpoint).",
    "Handles a 131,072-token context window, up to 32,768 output tokens.",
    "Open weights under Apache 2.0 — permissive licence.",
    "Training knowledge cutoff: September 2025.",
    "Scores 1303 on the LMArena (Chatbot Arena) (human-preference leaderboard) for qwen-2.5-72b.",
    "Cheapest of 2 providers serving qwen-2.5-72b: $0.76 blended per 1M tokens, vs $0.78 for the next cheapest.",
    "Value: 1303 Elo at $0.76 blended = 1714 Elo points per dollar — ranking #88 of 321 priced & rated endpoints in the index."
  ]
}