{
  "@context": "https://schema.org",
  "@type": "Dataset",
  "@id": "https://requesty.ai/data/finish-reason-mix-by-model-april-2026",
  "id": "finish-reason-by-model-april-2026",
  "slug": "finish-reason-mix-by-model-april-2026",
  "title": "finish_reason mix per model, April 2026",
  "shortTitle": "finish_reason by model",
  "topic": "agentic",
  "abstract": "Which AI models are used most for tool calling? In April 2026 Claude Opus 4.6 returned `finish_reason = tool_calls` 59% of the time on the Requesty gateway, the most agentic model on the platform. Gemini 2.5 Flash came second at 37%. Same-family Claude Sonnet 4.5 only 9%, and the entire OpenAI lineup (GPT-4o, GPT-4.1-mini, GPT-4.1-nano, GPT-5-mini) sat under 4%.",
  "whyItMatters": "Two models from the same provider can have completely different agentic profiles, which means choosing a frontier model for an agent based on brand alone is a coin flip. The headline \"Anthropic is agentic\" framing on the per-provider chart is really an Opus 4.6 effect: Sonnet 4.5 behaves more like a chat model in production traffic, despite both being marketed as agentic-capable.",
  "questions": [
    "Which AI models are used most for tool calling?",
    "Is Claude Opus more agentic than Claude Sonnet in production?",
    "Which OpenAI models do AI agents use?",
    "How agentic is Gemini 2.5 Flash compared to Claude?"
  ],
  "period": "Apr 2026",
  "updated": "2026-05-09",
  "license": "CC BY 4.0",
  "licenseUrl": "https://creativecommons.org/licenses/by/4.0/",
  "caveats": [
    "finish_reason was not populated before 2026, so this is April 2026 only.",
    "Aggregating finish_reason at the model level smooths over how the model is invoked. A model used inside an agent loop will show more tool_calls than the same model used in a one-shot chatbot."
  ],
  "keyFindings": [
    "claude-opus-4-6: 59% tool_calls. The single most agentic model on the platform.",
    "gemini-2.5-flash: 37% tool_calls. The mid-tier general-purpose model that is doing real agentic work.",
    "claude-sonnet-4-5: 9% tool_calls. The same provider, the same family, dramatically less agentic.",
    "OpenAI lineup (gpt-4o, gpt-4.1-mini, gpt-4.1-nano, gpt-5-mini): all under 4% tool_calls.",
    "Practical implication: the \"agentic provider\" framing on the per-provider chart is really an \"Opus 4.6 effect\". Anthropic-direct looks agentic because Opus is."
  ],
  "columns": [
    {
      "key": "model",
      "label": "Model",
      "unit": "count"
    },
    {
      "key": "tool_calls",
      "label": "tool_calls",
      "unit": "percent"
    },
    {
      "key": "stop",
      "label": "stop",
      "unit": "percent"
    },
    {
      "key": "length",
      "label": "length",
      "unit": "percent"
    }
  ],
  "rows": [
    {
      "model": "claude-opus-4-6",
      "tool_calls": 0.594,
      "stop": 0.395,
      "length": 0.011
    },
    {
      "model": "gemini-2.5-flash",
      "tool_calls": 0.366,
      "stop": 0.612,
      "length": 0.021
    },
    {
      "model": "claude-sonnet-4-5",
      "tool_calls": 0.091,
      "stop": 0.907,
      "length": 0.002
    },
    {
      "model": "gpt-5-mini",
      "tool_calls": 0.035,
      "stop": 0.94,
      "length": 0.024
    },
    {
      "model": "gpt-4o",
      "tool_calls": 0.002,
      "stop": 0.998,
      "length": 0
    },
    {
      "model": "gpt-4.1-mini",
      "tool_calls": 0.002,
      "stop": 0.998,
      "length": 0
    },
    {
      "model": "deepseek-chat",
      "tool_calls": 0.005,
      "stop": 0.972,
      "length": 0.023
    },
    {
      "model": "gpt-4.1-nano",
      "tool_calls": 0,
      "stop": 0.999,
      "length": 0
    },
    {
      "model": "gemini-2.5-flash-lite",
      "tool_calls": 0,
      "stop": 0.998,
      "length": 0.002
    },
    {
      "model": "grok-4-1-fast",
      "tool_calls": 0.001,
      "stop": 0.998,
      "length": 0.001
    }
  ],
  "rowKey": "model",
  "citation": {
    "apa": "Requesty (2026). finish_reason mix per model, April 2026. Requesty Data. https://requesty.ai/data/finish-reason-mix-by-model-april-2026",
    "bibtex": "@misc{requesty_finish_reason_mix_by_model_april_2026,\n  author       = {{Requesty}},\n  title        = {finish\\_reason mix per model, April 2026},\n  year         = {2026},\n  howpublished = {\\url{https://requesty.ai/data/finish-reason-mix-by-model-april-2026}},\n  note         = {Requesty Data}\n}"
  },
  "permalink": "https://requesty.ai/data/finish-reason-mix-by-model-april-2026",
  "downloads": {
    "json": "https://requesty.ai/data/finish-reason-mix-by-model-april-2026/data.json",
    "csv": "https://requesty.ai/data/finish-reason-mix-by-model-april-2026/data.csv",
    "markdown": "https://requesty.ai/data/finish-reason-mix-by-model-april-2026.md"
  },
  "citedIn": [],
  "image": "https://requesty.ai/data/finish-reason-mix-by-model-april-2026/opengraph-image",
  "source": {
    "organization": "Requesty",
    "url": "https://requesty.ai"
  }
}