GPT-5.5

{
  "id": "openai/gpt-5.5",
  "name": "GPT-5.5",
  "description": "Default frontier GPT for coding, computer use, research, and knowledge work",
  "family": "gpt",
  "last_updated": "2026-04-23",
  "knowledge": "2025-12-01",
  "attachment": true,
  "reasoning": true,
  "tool_call": true,
  "structured_output": true,
  "temperature": false,
  "modalities": {
    "input": [
      "text",
      "image",
      "pdf"
    ],
    "output": [
      "text"
    ]
  },
  "limit": {
    "context": 1050000,
    "input": 922000,
    "output": 128000
  },
  "benchmarks": [
    {
      "name": "SWE-Bench Pro",
      "score": 58.6,
      "source": "https://www.anthropic.com/news/claude-opus-4-8",
      "metric": "resolve rate",
      "date": "2026-05-28"
    },
    {
      "name": "Terminal-Bench",
      "score": 78.2,
      "source": "https://www.anthropic.com/news/claude-opus-4-8",
      "metric": "success rate",
      "harness": "Terminus-2",
      "version": "2.1",
      "date": "2026-05-28"
    },
    {
      "name": "SWE-Atlas Codebase QnA",
      "score": 45.43,
      "source": "https://labs.scale.com/leaderboard/sweatlas-qna",
      "metric": "score",
      "harness": "Codex"
    },
    {
      "name": "SWE-Atlas Refactoring",
      "score": 44.79,
      "source": "https://labs.scale.com/leaderboard/sweatlas-refactoring",
      "metric": "score",
      "harness": "Codex"
    },
    {
      "name": "SWE-Atlas Test Writing",
      "score": 42.59,
      "source": "https://labs.scale.com/leaderboard/sweatlas-tw",
      "metric": "score",
      "harness": "Codex"
    },
    {
      "name": "Artificial Analysis Coding Agent Index",
      "score": 65.3,
      "source": "https://artificialanalysis.ai/agents/coding-agents",
      "metric": "average pass@1",
      "harness": "Codex",
      "variant": "xhigh"
    },
    {
      "name": "SWE-Atlas Codebase QnA",
      "score": 80.8,
      "source": "https://artificialanalysis.ai/agents/coding-agents",
      "metric": "pass@1",
      "harness": "Codex",
      "variant": "xhigh"
    },
    {
      "name": "SWE-Bench Pro",
      "score": 30.9,
      "source": "https://artificialanalysis.ai/agents/coding-agents",
      "metric": "pass@1",
      "harness": "Codex",
      "variant": "xhigh",
      "dataset": "hard-aa"
    },
    {
      "name": "Terminal-Bench",
      "score": 84.1,
      "source": "https://artificialanalysis.ai/agents/coding-agents",
      "metric": "pass@1",
      "harness": "Codex",
      "variant": "xhigh",
      "version": "2.1"
    },
    {
      "name": "Artificial Analysis Coding Agent Index",
      "score": 60.4,
      "source": "https://artificialanalysis.ai/agents/coding-agents",
      "metric": "average pass@1",
      "harness": "Codex",
      "variant": "medium"
    },
    {
      "name": "SWE-Atlas Codebase QnA",
      "score": 79.1,
      "source": "https://artificialanalysis.ai/agents/coding-agents",
      "metric": "pass@1",
      "harness": "Codex",
      "variant": "medium"
    },
    {
      "name": "SWE-Bench Pro",
      "score": 26.2,
      "source": "https://artificialanalysis.ai/agents/coding-agents",
      "metric": "pass@1",
      "harness": "Codex",
      "variant": "medium",
      "dataset": "hard-aa"
    },
    {
      "name": "Terminal-Bench",
      "score": 75.8,
      "source": "https://artificialanalysis.ai/agents/coding-agents",
      "metric": "pass@1",
      "harness": "Codex",
      "variant": "medium",
      "version": "2.1"
    },
    {
      "name": "Artificial Analysis Coding Agent Index",
      "score": 57.8,
      "source": "https://artificialanalysis.ai/agents/coding-agents",
      "metric": "average pass@1",
      "harness": "Cursor CLI",
      "variant": "medium"
    },
    {
      "name": "SWE-Atlas Codebase QnA",
      "score": 75,
      "source": "https://artificialanalysis.ai/agents/coding-agents",
      "metric": "pass@1",
      "harness": "Cursor CLI",
      "variant": "medium"
    },
    {
      "name": "SWE-Bench Pro",
      "score": 24.9,
      "source": "https://artificialanalysis.ai/agents/coding-agents",
      "metric": "pass@1",
      "harness": "Cursor CLI",
      "variant": "medium",
      "dataset": "hard-aa"
    },
    {
      "name": "Terminal-Bench",
      "score": 73.4,
      "source": "https://artificialanalysis.ai/agents/coding-agents",
      "metric": "pass@1",
      "harness": "Cursor CLI",
      "variant": "medium",
      "version": "2.1"
    },
    {
      "name": "Terminal-Bench",
      "score": 82.7,
      "source": "https://openai.com/index/introducing-gpt-5-5/",
      "metric": "success rate",
      "version": "2.0",
      "date": "2026-04-23"
    },
    {
      "name": "GPQA Diamond",
      "score": 93.6,
      "source": "https://openai.com/index/introducing-gpt-5-5/",
      "metric": "accuracy",
      "date": "2026-04-23"
    },
    {
      "name": "Humanity's Last Exam",
      "score": 41.4,
      "source": "https://openai.com/index/introducing-gpt-5-5/",
      "metric": "accuracy",
      "variant": "no tools",
      "date": "2026-04-23"
    },
    {
      "name": "Humanity's Last Exam",
      "score": 52.2,
      "source": "https://openai.com/index/introducing-gpt-5-5/",
      "metric": "accuracy",
      "variant": "with tools",
      "date": "2026-04-23"
    },
    {
      "name": "OSWorld-Verified",
      "score": 78.7,
      "source": "https://openai.com/index/introducing-gpt-5-5/",
      "metric": "success rate",
      "date": "2026-04-23"
    },
    {
      "name": "BrowseComp",
      "score": 84.4,
      "source": "https://openai.com/index/introducing-gpt-5-5/",
      "metric": "accuracy",
      "date": "2026-04-23"
    },
    {
      "name": "MMMU Pro",
      "score": 81.2,
      "source": "https://openai.com/index/introducing-gpt-5-5/",
      "metric": "accuracy",
      "variant": "no tools",
      "date": "2026-04-23"
    },
    {
      "name": "ARC-AGI-2",
      "score": 85,
      "source": "https://openai.com/index/introducing-gpt-5-5/",
      "metric": "accuracy",
      "variant": "Verified",
      "date": "2026-04-23"
    },
    {
      "name": "FrontierMath",
      "score": 51.7,
      "source": "https://openai.com/index/introducing-gpt-5-5/",
      "metric": "accuracy",
      "dataset": "Tier 1-3",
      "date": "2026-04-23"
    },
    {
      "name": "FrontierMath",
      "score": 35.4,
      "source": "https://openai.com/index/introducing-gpt-5-5/",
      "metric": "accuracy",
      "dataset": "Tier 4",
      "date": "2026-04-23"
    },
    {
      "name": "GDPval",
      "score": 84.9,
      "source": "https://openai.com/index/introducing-gpt-5-5/",
      "metric": "wins or ties",
      "date": "2026-04-23"
    },
    {
      "name": "MCP Atlas",
      "score": 75.3,
      "source": "https://openai.com/index/introducing-gpt-5-5/",
      "metric": "success rate",
      "date": "2026-04-23"
    },
    {
      "name": "Toolathlon",
      "score": 55.6,
      "source": "https://openai.com/index/introducing-gpt-5-5/",
      "metric": "success rate",
      "date": "2026-04-23"
    },
    {
      "name": "τ²-Bench Telecom",
      "score": 98,
      "source": "https://openai.com/index/introducing-gpt-5-5/",
      "metric": "success rate",
      "variant": "original prompts",
      "date": "2026-04-23"
    }
  ],
  "release_date": "2026-04-23",
  "open_weights": false
}