{
  "data": {
    "area": "agent-tasks",
    "attribution": "τ²-bench, Sierra Research, MIT",
    "cadence": "as submissions land",
    "categories": [
      "inference",
      "open-weights"
    ],
    "ceiling": 100,
    "competitor": "",
    "data": "https://github.com/sierra-research/tau2-bench/tree/main/web/leaderboard/public/submissions",
    "direction": "higher",
    "floor": 0,
    "json": "https://www.anchorterminal.com/benchmarks/tau2-bench/index.json",
    "licence": "MIT",
    "licenceUrl": "https://github.com/sierra-research/tau2-bench/blob/main/LICENSE",
    "measures": "A model working as a customer-service agent under a written policy, with tools and a simulated customer, in one domain at a time (airline, retail, telecom, banking), in text or by voice. Each domain is scored on its own.",
    "method": "https://taubench.com/",
    "models": 6,
    "name": "τ²-bench",
    "newest": "2026-10-29",
    "page": "https://www.anchorterminal.com/benchmarks/tau2-bench/",
    "read": "2026-11-20",
    "results": 14,
    "runBy": "Sierra",
    "saturated": false,
    "slug": "tau2-bench",
    "status": "current",
    "unit": "% of tasks passed on the first try (pass^1)",
    "versions": [
      {
        "rows": [
          {
            "as": "Kimi K3",
            "date": "2026-08-04",
            "listing": "kimi-k3",
            "model": "kimi-k3",
            "name": "Kimi K3",
            "of": 4,
            "rank": 1,
            "runBy": "Sierra",
            "score": 37.11,
            "setting": {
              "effort": "max",
              "retrieval": "alltools",
              "submission": "kimi-k3_sierra_2026-08-04",
              "submissionType": "standard",
              "tau2Version": "1.0.1",
              "userSimulator": "gpt-5.2"
            },
            "tier": "independent",
            "url": "https://taubench.com/",
            "version": "text: banking_knowledge"
          },
          {
            "as": "Gemini 3.1 Pro Preview",
            "date": "2026-05-05",
            "listing": "vertex-ai",
            "model": "gemini-3-1-pro-preview",
            "name": "Gemini 3.1 Pro (preview)",
            "of": 4,
            "rank": 2,
            "runBy": "Sierra",
            "score": 26.03,
            "setting": {
              "effort": "high",
              "retrieval": "alltools",
              "submission": "gemini-3-1-pro-preview_sierra_2026-05-05",
              "submissionType": "standard",
              "tau2Version": "1.0.1",
              "userSimulator": "gpt-5.2"
            },
            "tier": "independent",
            "url": "https://taubench.com/",
            "version": "text: banking_knowledge"
          },
          {
            "as": "Inkling",
            "date": "2026-08-04",
            "listing": "inkling",
            "model": "inkling",
            "name": "Inkling",
            "of": 4,
            "rank": 3,
            "runBy": "Sierra",
            "score": 25,
            "setting": {
              "effort": "max",
              "retrieval": "alltools",
              "submission": "inkling_sierra_2026-08-04",
              "submissionType": "standard",
              "tau2Version": "1.0.1",
              "userSimulator": "gpt-5.2"
            },
            "tier": "independent",
            "url": "https://taubench.com/",
            "version": "text: banking_knowledge"
          },
          {
            "as": "Qwen3.5-397B-A17B",
            "date": "2026-03-02",
            "listing": "qwen3-5-397b-a17b",
            "model": "qwen3-5-397b-a17b",
            "name": "Qwen3.5-397B-A17B",
            "of": 4,
            "rank": 4,
            "runBy": "Sierra",
            "score": 9.79,
            "setting": {
              "effort": "enabled",
              "retrieval": "text-emb-3-large",
              "submission": "qwen3.5-397b-a17b-think_sierra_2026-03-02",
              "submissionType": "standard",
              "tau2Version": "1.0.1",
              "userSimulator": "gpt-5.2"
            },
            "tier": "independent",
            "url": "https://taubench.com/",
            "version": "text: banking_knowledge"
          }
        ],
        "version": "text: banking_knowledge"
      },
      {
        "rows": [
          {
            "as": "gpt-live-1",
            "date": "2026-09-16",
            "listing": "gpt-live",
            "model": "gpt-live",
            "name": "GPT-Live",
            "of": 1,
            "rank": 1,
            "runBy": "OpenAI",
            "score": 37.11,
            "setting": {
              "effort": "Backend: GPT-6 Astra (high)",
              "retrieval": "alltools",
              "submission": "gpt-live-1-astra-high-banking_openai_2026-09-16",
              "submissionType": "custom",
              "tau2Version": "1.0.1",
              "userSimulator": "gpt-5.5 (xhigh)"
            },
            "tier": "submitted",
            "url": "https://taubench.com/",
            "version": "voice: banking_knowledge"
          }
        ],
        "version": "voice: banking_knowledge"
      },
      {
        "rows": [
          {
            "as": "gpt-live-1",
            "date": "2026-10-20",
            "listing": "gpt-live",
            "model": "gpt-live",
            "name": "GPT-Live",
            "of": 1,
            "rank": 1,
            "runBy": "Sierra",
            "score": 60.33,
            "setting": {
              "agentProvider": "openai_live",
              "submission": "gpt-live-1_openai_2026-10-20_entity",
              "submissionType": "standard"
            },
            "tier": "independent",
            "url": "https://taubench.com/",
            "version": "voice-v2-entity"
          }
        ],
        "version": "voice-v2-entity"
      },
      {
        "rows": [
          {
            "as": "gpt-live-1",
            "date": "2026-10-20",
            "listing": "gpt-live",
            "model": "gpt-live",
            "name": "GPT-Live",
            "of": 1,
            "rank": 1,
            "runBy": "Sierra",
            "score": 33.16,
            "setting": {
              "agentProvider": "openai_live",
              "submission": "gpt-live-1_openai_2026-10-20_banking",
              "submissionType": "standard"
            },
            "tier": "independent",
            "url": "https://taubench.com/",
            "version": "voice-v2-banking"
          }
        ],
        "version": "voice-v2-banking"
      },
      {
        "rows": [
          {
            "as": "gpt-live-1",
            "date": "2026-10-20",
            "listing": "gpt-live",
            "model": "gpt-live",
            "name": "GPT-Live",
            "of": 1,
            "rank": 1,
            "runBy": "Sierra",
            "score": 79.91,
            "setting": {
              "agentProvider": "openai_live",
              "submission": "gpt-live-1_openai_2026-10-20",
              "submissionType": "standard"
            },
            "tier": "independent",
            "url": "https://taubench.com/",
            "version": "voice-v2"
          }
        ],
        "version": "voice-v2"
      },
      {
        "rows": [
          {
            "as": "DeepSeek-V3.2",
            "date": "2025-12-01",
            "listing": "deepseek-v3-2",
            "model": "deepseek-v3-2",
            "name": "DeepSeek-V3.2",
            "of": 1,
            "rank": 1,
            "runBy": "DeepSeek",
            "score": 96.2,
            "setting": {
              "submission": "deepseek-v3.2_deepseek_2025-12-01",
              "tau2Version": "v0.1.3",
              "userSimulator": "DeepSeek-V3.2"
            },
            "tier": "submitted",
            "url": "https://taubench.com/",
            "version": "text: telecom (legacy)"
          }
        ],
        "version": "text: telecom (legacy)"
      },
      {
        "rows": [
          {
            "as": "Qwen3.5-397B-A17B",
            "date": "2026-03-02",
            "listing": "qwen3-5-397b-a17b",
            "model": "qwen3-5-397b-a17b",
            "name": "Qwen3.5-397B-A17B",
            "of": 1,
            "rank": 1,
            "runBy": "Sierra",
            "score": 97.81,
            "setting": {
              "effort": "enabled",
              "submission": "qwen3.5-397b-a17b-think_sierra_2026-03-02",
              "submissionType": "standard",
              "tau2Version": "1.0.1",
              "userSimulator": "gpt-5.2"
            },
            "tier": "independent",
            "url": "https://taubench.com/",
            "version": "text: telecom"
          }
        ],
        "version": "text: telecom"
      },
      {
        "rows": [
          {
            "as": "DeepSeek-V3.2",
            "date": "2025-12-01",
            "listing": "deepseek-v3-2",
            "model": "deepseek-v3-2",
            "name": "DeepSeek-V3.2",
            "of": 1,
            "rank": 1,
            "runBy": "DeepSeek",
            "score": 81.1,
            "setting": {
              "submission": "deepseek-v3.2_deepseek_2025-12-01",
              "tau2Version": "v0.1.3",
              "userSimulator": "DeepSeek-V3.2"
            },
            "tier": "submitted",
            "url": "https://taubench.com/",
            "version": "text: retail (legacy)"
          }
        ],
        "version": "text: retail (legacy)"
      },
      {
        "rows": [
          {
            "as": "Qwen3.5-397B-A17B",
            "date": "2026-03-02",
            "listing": "qwen3-5-397b-a17b",
            "model": "qwen3-5-397b-a17b",
            "name": "Qwen3.5-397B-A17B",
            "of": 1,
            "rank": 1,
            "runBy": "Sierra",
            "score": 84.43,
            "setting": {
              "effort": "enabled",
              "submission": "qwen3.5-397b-a17b-think_sierra_2026-03-02",
              "submissionType": "standard",
              "tau2Version": "1.0.1",
              "userSimulator": "gpt-5.2"
            },
            "tier": "independent",
            "url": "https://taubench.com/",
            "version": "text: retail"
          }
        ],
        "version": "text: retail"
      },
      {
        "rows": [
          {
            "as": "DeepSeek-V3.2",
            "date": "2025-12-01",
            "listing": "deepseek-v3-2",
            "model": "deepseek-v3-2",
            "name": "DeepSeek-V3.2",
            "of": 1,
            "rank": 1,
            "runBy": "DeepSeek",
            "score": 63.8,
            "setting": {
              "submission": "deepseek-v3.2_deepseek_2025-12-01",
              "tau2Version": "v0.1.3",
              "userSimulator": "DeepSeek-V3.2"
            },
            "tier": "submitted",
            "url": "https://taubench.com/",
            "version": "text: airline (legacy)"
          }
        ],
        "version": "text: airline (legacy)"
      },
      {
        "rows": [
          {
            "as": "Qwen3.5-397B-A17B",
            "date": "2026-03-02",
            "listing": "qwen3-5-397b-a17b",
            "model": "qwen3-5-397b-a17b",
            "name": "Qwen3.5-397B-A17B",
            "of": 1,
            "rank": 1,
            "runBy": "Sierra",
            "score": 81.5,
            "setting": {
              "effort": "enabled",
              "submission": "qwen3.5-397b-a17b-think_sierra_2026-03-02",
              "submissionType": "standard",
              "tau2Version": "1.0.1",
              "userSimulator": "gpt-5.2"
            },
            "tier": "independent",
            "url": "https://taubench.com/",
            "version": "text: airline"
          }
        ],
        "version": "text: airline"
      }
    ]
  },
  "kind": "anchor.page",
  "links": {
    "api": "https://www.anchorterminal.com/api/v1/index.json",
    "html": "https://www.anchorterminal.com/benchmarks/tau2-bench/",
    "json": "https://www.anchorterminal.com/benchmarks/tau2-bench/index.json",
    "llms": "https://www.anchorterminal.com/llms.txt",
    "markdown": "https://www.anchorterminal.com/benchmarks/tau2-bench/index.md",
    "mcp": "https://www.anchorterminal.com/mcp",
    "slim": "https://www.anchorterminal.com/benchmarks/tau2-bench/index.min.md"
  },
  "markdown": "A model working as a customer-service agent under a written policy, with tools and a simulated customer, in one domain at a time (airline, retail, telecom, banking), in text or by voice. Each domain is scored on its own.\n\n- Score: % of tasks passed on the first try (pass^1), higher is better\n- Run by: Sierra\n- Status: Current, newest result 29 October 2026\n- Updated: as submissions land\n- Data: https://github.com/sierra-research/tau2-bench/tree/main/web/leaderboard/public/submissions (MIT, https://github.com/sierra-research/tau2-bench/blob/main/LICENSE), read 2026-11-20\n- Method: https://taubench.com/\n\n## Results: text: banking knowledge\n\n4 models we list, each at its best setting, best first.\n\n| # | Model | Score | Range | Who ran it | Setting | Date | Source |\n| --- | --- | --- | --- | --- | --- | --- | --- |\n| 1 | [Kimi K3](https://www.anchorterminal.com/tools/kimi-k3.md) | 37.1% |  | Independent | effort max, retrieval alltools, submission type standard, tau2 version 1.0.1, user simulator gpt-5.2 | 2026-08-04 | https://taubench.com/ |\n| 2 | [Gemini 3.1 Pro (preview)](https://www.anchorterminal.com/tools/vertex-ai.md) | 26.0% |  | Independent | effort high, retrieval alltools, submission type standard, tau2 version 1.0.1, user simulator gpt-5.2 | 2026-05-05 | https://taubench.com/ |\n| 3 | [Inkling](https://www.anchorterminal.com/tools/inkling.md) | 25.0% |  | Independent | effort max, retrieval alltools, submission type standard, tau2 version 1.0.1, user simulator gpt-5.2 | 2026-08-04 | https://taubench.com/ |\n| 4 | [Qwen3.5-397B-A17B](https://www.anchorterminal.com/tools/qwen3-5-397b-a17b.md) | 9.8% |  | Independent | effort enabled, retrieval text-emb-3-large, submission type standard, tau2 version 1.0.1, user simulator gpt-5.2 | 2026-03-02 | https://taubench.com/ |\n\n## Results: voice: banking knowledge\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 | [GPT-Live](https://www.anchorterminal.com/tools/gpt-live.md) | 37.1% |  | Submitted by the maker | effort Backend: GPT-6 Astra (high), retrieval alltools, submission type custom, tau2 version 1.0.1, user simulator gpt-5.5 (xhigh) | 2026-09-16 | https://taubench.com/ |\n\n## Results: voice-v2-entity\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 | [GPT-Live](https://www.anchorterminal.com/tools/gpt-live.md) | 60.3% |  | Independent | agent provider openai_live, submission type standard | 2026-10-20 | https://taubench.com/ |\n\n## Results: voice-v2-banking\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 | [GPT-Live](https://www.anchorterminal.com/tools/gpt-live.md) | 33.2% |  | Independent | agent provider openai_live, submission type standard | 2026-10-20 | https://taubench.com/ |\n\n## Results: voice-v2\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 | [GPT-Live](https://www.anchorterminal.com/tools/gpt-live.md) | 79.9% |  | Independent | agent provider openai_live, submission type standard | 2026-10-20 | https://taubench.com/ |\n\n## Results: text: telecom (legacy)\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-V3.2](https://www.anchorterminal.com/tools/deepseek-v3-2.md) | 96.2% |  | Submitted by the maker | tau2 version v0.1.3, user simulator DeepSeek-V3.2 | 2025-12-01 | https://taubench.com/ |\n\n## Results: text: telecom\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 | [Qwen3.5-397B-A17B](https://www.anchorterminal.com/tools/qwen3-5-397b-a17b.md) | 97.8% |  | Independent | effort enabled, submission type standard, tau2 version 1.0.1, user simulator gpt-5.2 | 2026-03-02 | https://taubench.com/ |\n\n## Results: text: retail (legacy)\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-V3.2](https://www.anchorterminal.com/tools/deepseek-v3-2.md) | 81.1% |  | Submitted by the maker | tau2 version v0.1.3, user simulator DeepSeek-V3.2 | 2025-12-01 | https://taubench.com/ |\n\n## Results: text: retail\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 | [Qwen3.5-397B-A17B](https://www.anchorterminal.com/tools/qwen3-5-397b-a17b.md) | 84.4% |  | Independent | effort enabled, submission type standard, tau2 version 1.0.1, user simulator gpt-5.2 | 2026-03-02 | https://taubench.com/ |\n\n## Results: text: airline (legacy)\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-V3.2](https://www.anchorterminal.com/tools/deepseek-v3-2.md) | 63.8% |  | Submitted by the maker | tau2 version v0.1.3, user simulator DeepSeek-V3.2 | 2025-12-01 | https://taubench.com/ |\n\n## Results: text: airline\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 | [Qwen3.5-397B-A17B](https://www.anchorterminal.com/tools/qwen3-5-397b-a17b.md) | 81.5% |  | Independent | effort enabled, submission type standard, tau2 version 1.0.1, user simulator gpt-5.2 | 2026-03-02 | https://taubench.com/ |\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: τ²-bench, Sierra Research, MIT, MIT. 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": "τ²-bench",
        "url": ""
      }
    ],
    "description": "τ²-bench results for 6 AI models we list, run by Sierra, with each result's setting, date and source.",
    "facts": [
      "6 models",
      "Sierra",
      "Current"
    ],
    "h1": "τ²-bench",
    "image": "https://www.anchorterminal.com/assets/og/benchmarks-tau2-bench.png",
    "path": "/benchmarks/tau2-bench/",
    "published": "",
    "section": "benchmarks",
    "title": "τ²-bench results for the AI models we list | Anchor Terminal",
    "toc": null,
    "updated": "2026-11-21",
    "url": "https://www.anchorterminal.com/benchmarks/tau2-bench/"
  },
  "tokens": {
    "markdown": 1750,
    "slim": 1630
  },
  "version": 1
}
