{
  "data": {
    "area": "coding",
    "attribution": "Epoch AI, Capabilities and benchmarking (CC BY 4.0)",
    "cadence": "as models are released",
    "categories": [
      "inference",
      "open-weights"
    ],
    "ceiling": 100,
    "competitor": "",
    "data": "https://epoch.ai/data/benchmark_data.zip",
    "direction": "higher",
    "floor": 0,
    "json": "https://www.anchorterminal.com/benchmarks/swe-bench-verified/index.json",
    "licence": "CC BY 4.0",
    "licenceUrl": "https://epoch.ai/benchmarks/use-this-data",
    "measures": "A model given a Python repository and a GitHub issue must change the code until the repository's own tests pass. Epoch AI runs 484 of the 500 human-checked issues in its own simple tool-calling scaffold, with no network, after a major scaffold upgrade in February 2026.",
    "method": "https://epoch.ai/benchmarks/swe-bench-verified",
    "models": 1,
    "name": "SWE-bench Verified",
    "newest": "2026-06-25",
    "page": "https://www.anchorterminal.com/benchmarks/swe-bench-verified/",
    "read": "2026-11-20",
    "results": 1,
    "runBy": "Epoch AI",
    "saturated": false,
    "slug": "swe-bench-verified",
    "status": "current",
    "unit": "% of issues resolved",
    "versions": [
      {
        "rows": [
          {
            "as": "deepseek-v4-pro",
            "date": "2026-06-18",
            "listing": "deepseek-api",
            "model": "deepseek-v4-pro",
            "name": "DeepSeek V4 Pro",
            "of": 1,
            "rank": 1,
            "runBy": "Epoch AI",
            "score": 77.64,
            "setting": {
              "effort": "max"
            },
            "tier": "independent",
            "url": "https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FWUk3f44JhxtAJaSxi6gFZR.eval"
          }
        ],
        "version": ""
      }
    ]
  },
  "kind": "anchor.page",
  "links": {
    "api": "https://www.anchorterminal.com/api/v1/index.json",
    "html": "https://www.anchorterminal.com/benchmarks/swe-bench-verified/",
    "json": "https://www.anchorterminal.com/benchmarks/swe-bench-verified/index.json",
    "llms": "https://www.anchorterminal.com/llms.txt",
    "markdown": "https://www.anchorterminal.com/benchmarks/swe-bench-verified/index.md",
    "mcp": "https://www.anchorterminal.com/mcp",
    "slim": "https://www.anchorterminal.com/benchmarks/swe-bench-verified/index.min.md"
  },
  "markdown": "A model given a Python repository and a GitHub issue must change the code until the repository's own tests pass. Epoch AI runs 484 of the 500 human-checked issues in its own simple tool-calling scaffold, with no network, after a major scaffold upgrade in February 2026.\n\n- Score: % of issues resolved, higher is better\n- Run by: Epoch AI\n- Status: Current, newest result 25 June 2026\n- Updated: as models are released\n- Data: https://epoch.ai/data/benchmark_data.zip (CC BY 4.0, https://epoch.ai/benchmarks/use-this-data), read 2026-11-20\n- Method: https://epoch.ai/benchmarks/swe-bench-verified\n\n## Results\n\n1 model we list, each at its best setting, best first.\n\n| # | Model | Score | Range | Who ran it | Setting | Date | Source |\n| --- | --- | --- | --- | --- | --- | --- | --- |\n| 1 | [DeepSeek V4 Pro](https://www.anchorterminal.com/tools/deepseek-api.md) | 77.6% |  | Independent | effort max | 2026-06-18 | https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FWUk3f44JhxtAJaSxi6gFZR.eval |\n\nA rank is among the models we list with a result, not a place on the whole leaderboard, and a model offered by several hosts is shown once, on its own listing. These results don't move our grades.\n\nData: Epoch AI, Capabilities and benchmarking (CC BY 4.0), CC BY 4.0. All benchmarks: https://www.anchorterminal.com/benchmarks/index.md\n",
  "meta": {
    "attribution": "Anchor Terminal (https://www.anchorterminal.com)",
    "attributionRequired": true,
    "credit": "CC BY 4.0, credit Anchor Terminal and link the page or API address you took it from (https://www.anchorterminal.com/terms/#data)",
    "docs": "https://www.anchorterminal.com/docs/",
    "generatedAt": "2026-11-22",
    "license": "CC-BY-4.0",
    "method": "https://www.anchorterminal.com/benchmark/",
    "methodology": "1.0",
    "openapi": "https://www.anchorterminal.com/openapi.json",
    "preview": false,
    "run": "2026-10-01",
    "runLabel": "October 2026 research run",
    "terms": "https://www.anchorterminal.com/terms/#data"
  },
  "page": {
    "breadcrumbs": [
      {
        "name": "Home",
        "url": "https://www.anchorterminal.com/"
      },
      {
        "name": "Benchmarks",
        "url": "https://www.anchorterminal.com/benchmarks/"
      },
      {
        "name": "SWE-bench Verified",
        "url": ""
      }
    ],
    "description": "SWE-bench Verified results for 1 AI models we list, run by Epoch AI, with each result's setting, date and source.",
    "facts": [
      "1 models",
      "Epoch AI",
      "Current"
    ],
    "h1": "SWE-bench Verified",
    "image": "https://www.anchorterminal.com/assets/og/benchmarks-swe-bench-verified.png",
    "path": "/benchmarks/swe-bench-verified/",
    "published": "",
    "section": "benchmarks",
    "title": "SWE-bench Verified results for the AI models we list | Anchor Terminal",
    "toc": null,
    "updated": "2026-11-21",
    "url": "https://www.anchorterminal.com/benchmarks/swe-bench-verified/"
  },
  "tokens": {
    "markdown": 500,
    "slim": 430
  },
  "version": 1
}
