{
  "@context": "https://schema.org",
  "@type": "Dataset",
  "@id": "https://www.requesty.ai/data/reasoning-token-share-by-provider-october-2026",
  "id": "reasoning-share-october-2026",
  "slug": "reasoning-token-share-by-provider-october-2026",
  "title": "Reasoning-token share of provider output, October 2026",
  "shortTitle": "Reasoning-token share",
  "topic": "agentic",
  "abstract": "How much provider output is reasoning? In October 2026 reasoning tokens made up 78% of output on Fireworks, the highest share, and 6% on Mistral, the lowest among providers with meaningful reasoning volume.",
  "whyItMatters": "Reasoning tokens are billed as output but never shown to users. Providers with a high share serve mostly thinking models, so output cost per visible answer is higher there.",
  "questions": [
    "Which providers serve the most reasoning tokens?",
    "What share of output tokens are reasoning tokens?",
    "How much do thinking models change output cost?"
  ],
  "period": "Oct 2026",
  "updated": "2026-11-11",
  "license": "CC BY 4.0",
  "licenseUrl": "https://creativecommons.org/licenses/by/4.0/",
  "caveats": [
    "reasoning_tokens divided by output_tokens. Providers with negligible reasoning output are omitted."
  ],
  "keyFindings": [
    "Fireworks: 78%.",
    "Melious: 69%.",
    "Vertex (Claude): 58%.",
    "TensorX: 56%.",
    "OpenAI: 48%."
  ],
  "columns": [
    {
      "key": "provider",
      "label": "Provider",
      "unit": "count"
    },
    {
      "key": "reasoning_share",
      "label": "Reasoning share",
      "unit": "percent"
    }
  ],
  "rows": [
    {
      "provider": "Vertex (Gemini)",
      "reasoning_share": 0.4079
    },
    {
      "provider": "Azure",
      "reasoning_share": 0.4182
    },
    {
      "provider": "TensorX",
      "reasoning_share": 0.5594
    },
    {
      "provider": "OpenAI",
      "reasoning_share": 0.4769
    },
    {
      "provider": "DeepInfra",
      "reasoning_share": 0.1005
    },
    {
      "provider": "Nebius",
      "reasoning_share": 0.1186
    },
    {
      "provider": "Bedrock",
      "reasoning_share": 0.2693
    },
    {
      "provider": "Anthropic",
      "reasoning_share": 0.4519
    },
    {
      "provider": "DeepSeek",
      "reasoning_share": 0.4546
    },
    {
      "provider": "Vertex (Claude)",
      "reasoning_share": 0.58
    },
    {
      "provider": "Novita",
      "reasoning_share": 0.2393
    },
    {
      "provider": "Melious",
      "reasoning_share": 0.6853
    },
    {
      "provider": "Mistral",
      "reasoning_share": 0.0646
    },
    {
      "provider": "Lyceum",
      "reasoning_share": 0.1174
    },
    {
      "provider": "Fireworks",
      "reasoning_share": 0.7849
    }
  ],
  "rowKey": "provider",
  "citation": {
    "apa": "Requesty (2026). Reasoning-token share of provider output, October 2026. Requesty Data. https://www.requesty.ai/data/reasoning-token-share-by-provider-october-2026",
    "bibtex": "@misc{requesty_reasoning_token_share_by_provider_october_2026,\n  author       = {{Requesty}},\n  title        = {Reasoning-token share of provider output, October 2026},\n  year         = {2026},\n  howpublished = {\\url{https://www.requesty.ai/data/reasoning-token-share-by-provider-october-2026}},\n  note         = {Requesty Data}\n}"
  },
  "permalink": "https://www.requesty.ai/data/reasoning-token-share-by-provider-october-2026",
  "downloads": {
    "json": "https://www.requesty.ai/data/reasoning-token-share-by-provider-october-2026/data.json",
    "csv": "https://www.requesty.ai/data/reasoning-token-share-by-provider-october-2026/data.csv",
    "markdown": "https://www.requesty.ai/data/reasoning-token-share-by-provider-october-2026.md"
  },
  "citedIn": [],
  "image": "https://www.requesty.ai/data/reasoning-token-share-by-provider-october-2026/opengraph-image",
  "source": {
    "organization": "Requesty",
    "url": "https://www.requesty.ai"
  }
}