{
  "name": "BenchSmoke AI model leaderboard",
  "url": "https://benchsmoke.com/leaderboard",
  "json_url": "https://benchsmoke.com/leaderboard.json",
  "last_updated": "2026-10-17",
  "dates_are": "calendar days in America/New_York",
  "read_this_first": [
    "Two separate sections. own_tests is what BenchSmoke tested hands-on. public_benchmarks is data published by other groups, each with source, link and retrieved date.",
    "There is no combined score and no overall rank. Do not add or average the two.",
    "own_tests.status \"not tested\" means BenchSmoke has not tested that model. It is not a score.",
    "Public figures belong to their publishers; see each source entry for its license. Cite BenchSmoke and the original source.",
    "This file contains only the free tier. Fuller run details are not published here."
  ],
  "own_tests_suite": {
    "version": "v1",
    "status": "designed and frozen on 2026-10-17; no model has been run yet",
    "runs_per_task": 3,
    "run_rules": [
      "Same prompt, same fixture, same harness for every model; prompt files below are hashed and any change means suite v2 and a full re-run.",
      "Each model is run through its provider's own API or through OpenRouter (the route is recorded per run). Reasoning effort is set to the highest named setting the provider offers and recorded. Other settings are left at the provider default and recorded.",
      "Every run is kept: all 3 runs per task are published, not the best one. A run that crashes, times out (20 minutes) or hits a rate limit is recorded as a fail with the reason, unless the cause was our harness, in which case it is re-run and the log says so.",
      "Scoring is automatic where possible (scripts saved with the evidence); any manual check is written down and done blind to which model produced the output.",
      "Results are shown per task as outcomes of the 3 runs (for example '2 pass, 1 partial'). There is no combined score and no rank across tasks."
    ],
    "outcome_rules": {
      "pass": "all acceptance checks passed",
      "partial": "at least half of the checks passed but not all",
      "fail": "fewer than half of the checks passed, or the run did not finish"
    },
    "tasks": [
      {
        "id": "T1",
        "title": "Small to-do app",
        "kind": "coding: build from scratch",
        "checks_total": 7,
        "prompt_sha256": "1825ff50387939c855606bdcdb95bf55a7fd4ec56a45db3efbfb607a46cc1b22",
        "prompt_url": "https://benchsmoke.com/leaderboard/suite/T1-todo-app.txt"
      },
      {
        "id": "T2",
        "title": "Fix a bug so the tests pass",
        "kind": "coding: debug",
        "checks_total": 5,
        "prompt_sha256": "96a5e9655a2c9ac3cf60456d38324822eae3ffe7b48c119101bf9d67042ddde2",
        "prompt_url": "https://benchsmoke.com/leaderboard/suite/T2-fix-failing-tests.txt"
      },
      {
        "id": "T3",
        "title": "Add search and pagination to an existing app",
        "kind": "coding: extend",
        "checks_total": 5,
        "prompt_sha256": "03c13c6d7d24ac4ec9d13c0735fcff7904a780c3b0326ecc1b69f7fbb95f362d",
        "prompt_url": "https://benchsmoke.com/leaderboard/suite/T3-add-feature.txt"
      },
      {
        "id": "T4",
        "title": "Landing page from a brief",
        "kind": "design and front end",
        "checks_total": 6,
        "prompt_sha256": "66ef2143b7ee3ad5b05c56ba40c97f1268cc7231fb7221c474e273c70e30bfbc",
        "prompt_url": "https://benchsmoke.com/leaderboard/suite/T4-landing-page.txt"
      },
      {
        "id": "T5",
        "title": "Three-step workflow with tools",
        "kind": "agent: tool use",
        "checks_total": 6,
        "prompt_sha256": "fe7af2cd9c449d8caed880b83af4acfef4b67d16e21d79a87977f4120d30fe30",
        "prompt_url": "https://benchsmoke.com/leaderboard/suite/T5-three-step-workflow.txt"
      }
    ]
  },
  "sources": {
    "lmarena_webdev": {
      "name": "LMArena WebDev Arena (overall, latest snapshot)",
      "publisher": "LMArena",
      "site_url": "https://lmarena.ai/",
      "data_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset/viewer/webdev/latest",
      "license": "CC-BY-4.0",
      "license_url": "https://creativecommons.org/licenses/by/4.0/",
      "terms_note": "Open data: the lmarena-ai/leaderboard-dataset card on Hugging Face declares CC-BY-4.0. We cite name, link and date, show a limited set of figures, and do not republish the full table.",
      "what_it_measures": "Human preference votes on web-app building tasks, shown as an Elo-style score with a rank.",
      "retrieved": "2026-10-17",
      "source_published": "2026-10-16",
      "dataset_revision": "0ebb411abf001fa30ad48fc7e49c5cd5891d5f8a",
      "rows_in_overall": 144,
      "status": "ok"
    },
    "lmarena_text": {
      "name": "LMArena Text Arena (overall, latest snapshot)",
      "publisher": "LMArena",
      "site_url": "https://lmarena.ai/",
      "data_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset/viewer/text/latest",
      "license": "CC-BY-4.0",
      "license_url": "https://creativecommons.org/licenses/by/4.0/",
      "terms_note": "Open data: the lmarena-ai/leaderboard-dataset card on Hugging Face declares CC-BY-4.0. We cite name, link and date, show a limited set of figures, and do not republish the full table.",
      "what_it_measures": "Human preference votes on general chat answers, shown as an Elo-style score with a rank. Not a coding test.",
      "retrieved": "2026-10-17",
      "source_published": "2026-10-15",
      "dataset_revision": "0ebb411abf001fa30ad48fc7e49c5cd5891d5f8a",
      "rows_in_overall": 416,
      "status": "ok"
    },
    "openrouter": {
      "name": "OpenRouter model list (list price and context window)",
      "publisher": "OpenRouter",
      "site_url": "https://openrouter.ai/models",
      "data_url": "https://openrouter.ai/api/v1/models",
      "license": "No open license. Public facts only (price, context length), cited and linked.",
      "license_url": "https://openrouter.ai/terms",
      "terms_note": "We call the documented public models endpoint once per refresh and keep only price and context length. OpenRouter prices can differ from the maker's own list price and can vary by time of day or provider. We do not scrape OpenRouter web pages.",
      "what_it_measures": "What OpenRouter lists as the base price per million tokens and the context window. Not a quality measure.",
      "retrieved": null,
      "source_published": null,
      "rows_in_overall": 471,
      "status": "Not published for now. OpenRouter terms of service are unresolved, so we show no OpenRouter figures. Vendor list prices will be added by hand, with a link to each vendor pricing page."
    },
    "swebench": {
      "name": "SWE-bench leaderboards",
      "publisher": "SWE-bench team",
      "site_url": "https://www.swebench.com/",
      "license": "CC BY-NC 4.0 (site data repo)",
      "license_url": "https://github.com/SWE-bench/swe-bench.github.io/blob/master/LICENSE",
      "status": "not used: the data is NonCommercial and BenchSmoke has sponsors and a planned paid tier. Link only, unless the maintainers give written permission.",
      "retrieved": null
    },
    "aider_polyglot": {
      "name": "Aider polyglot leaderboard",
      "publisher": "Aider",
      "site_url": "https://aider.chat/docs/leaderboards/",
      "license": "Apache-2.0 (repo data file)",
      "license_url": "https://github.com/Aider-AI/aider/blob/main/LICENSE.txt",
      "status": "not used yet: the public data file stops at 2025-10-03 and has none of our 22 models. The refresh routine re-checks it.",
      "retrieved": null,
      "latest_entry_date_seen": "2025-10-03",
      "checked": "2026-10-17"
    },
    "artificial_analysis": {
      "name": "Artificial Analysis free API",
      "publisher": "Artificial Analysis",
      "site_url": "https://artificialanalysis.ai/",
      "license": "Free API with required attribution; account and API key needed",
      "license_url": "https://artificialanalysis.ai/api-reference",
      "status": "not used yet: needs a free account and key that Justin has not created. Responds 401 without a key.",
      "retrieved": null
    }
  },
  "models": [
    {
      "id": "qwen3-8-27b",
      "name": "Qwen3.8 27B",
      "vendor": "Alibaba (Qwen)",
      "license": "Apache 2.0",
      "access": "open-weight-or-open-license",
      "own_tests": {
        "status": "not tested",
        "verdict": null,
        "tasks": {
          "T1": {
            "status": "not tested"
          },
          "T2": {
            "status": "not tested"
          },
          "T3": {
            "status": "not tested"
          },
          "T4": {
            "status": "not tested"
          },
          "T5": {
            "status": "not tested"
          }
        }
      },
      "public_benchmarks": {
        "lmarena_text": {
          "arena_score": {
            "value": 1441,
            "unit": "Elo-style points",
            "rank": 75,
            "sample_size": 23294,
            "ci_low": 1435,
            "ci_high": 1446,
            "source_entry_name": "qwen3.8-27b",
            "url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset/viewer/text/latest",
            "retrieved": "2026-10-17",
            "source_date": "2026-10-15"
          }
        },
        "lmarena_webdev": {
          "arena_score": {
            "value": 1593,
            "unit": "Elo-style points",
            "rank": 29,
            "sample_size": 15940,
            "ci_low": 1587,
            "ci_high": 1600,
            "source_entry_name": "Qwen3.8-27B",
            "url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset/viewer/webdev/latest",
            "retrieved": "2026-10-17",
            "source_date": "2026-10-16"
          }
        }
      }
    },
    {
      "id": "qwen3-8-max-0902",
      "name": "Qwen3.8 Max (0902)",
      "vendor": "Alibaba (Qwen)",
      "license": "Proprietary",
      "access": "closed",
      "own_tests": {
        "status": "not tested",
        "verdict": null,
        "tasks": {
          "T1": {
            "status": "not tested"
          },
          "T2": {
            "status": "not tested"
          },
          "T3": {
            "status": "not tested"
          },
          "T4": {
            "status": "not tested"
          },
          "T5": {
            "status": "not tested"
          }
        }
      },
      "public_benchmarks": {
        "lmarena_webdev": {
          "arena_score": {
            "value": 1674,
            "unit": "Elo-style points",
            "rank": 10,
            "sample_size": 13812,
            "ci_low": 1666,
            "ci_high": 1681,
            "source_entry_name": "Qwen3.8-Max (0902)",
            "url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset/viewer/webdev/latest",
            "retrieved": "2026-10-17",
            "source_date": "2026-10-16"
          }
        }
      }
    },
    {
      "id": "claude-fable-5-1",
      "name": "Claude Fable 5.1",
      "vendor": "Anthropic",
      "license": "Proprietary",
      "access": "closed",
      "own_tests": {
        "status": "not tested",
        "verdict": null,
        "tasks": {
          "T1": {
            "status": "not tested"
          },
          "T2": {
            "status": "not tested"
          },
          "T3": {
            "status": "not tested"
          },
          "T4": {
            "status": "not tested"
          },
          "T5": {
            "status": "not tested"
          }
        }
      },
      "public_benchmarks": {
        "lmarena_text": {
          "arena_score": {
            "value": 1512,
            "unit": "Elo-style points",
            "rank": 3,
            "sample_size": 14590,
            "ci_low": 1506,
            "ci_high": 1518,
            "source_entry_name": "Claude Fable 5.1 (Max)",
            "url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset/viewer/text/latest",
            "retrieved": "2026-10-17",
            "source_date": "2026-10-15"
          }
        },
        "lmarena_webdev": {
          "arena_score": {
            "value": 1741,
            "unit": "Elo-style points",
            "rank": 5,
            "sample_size": 7503,
            "ci_low": 1732,
            "ci_high": 1750,
            "source_entry_name": "Claude Fable 5.1 (Max)",
            "url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset/viewer/webdev/latest",
            "retrieved": "2026-10-17",
            "source_date": "2026-10-16"
          }
        }
      }
    },
    {
      "id": "claude-haiku-5-5",
      "name": "Claude Haiku 5.5",
      "vendor": "Anthropic",
      "license": "Proprietary",
      "access": "closed",
      "own_tests": {
        "status": "not tested",
        "verdict": null,
        "tasks": {
          "T1": {
            "status": "not tested"
          },
          "T2": {
            "status": "not tested"
          },
          "T3": {
            "status": "not tested"
          },
          "T4": {
            "status": "not tested"
          },
          "T5": {
            "status": "not tested"
          }
        }
      },
      "public_benchmarks": {
        "lmarena_webdev": {
          "arena_score": {
            "value": 1585,
            "unit": "Elo-style points",
            "rank": 32,
            "sample_size": 3156,
            "ci_low": 1573,
            "ci_high": 1596,
            "source_entry_name": "Claude Haiku 5.5 (High)",
            "url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset/viewer/webdev/latest",
            "retrieved": "2026-10-17",
            "source_date": "2026-10-16"
          }
        }
      }
    },
    {
      "id": "claude-opus-5-5",
      "name": "Claude Opus 5.5",
      "vendor": "Anthropic",
      "license": "Proprietary",
      "access": "closed",
      "own_tests": {
        "status": "not tested",
        "verdict": null,
        "tasks": {
          "T1": {
            "status": "not tested"
          },
          "T2": {
            "status": "not tested"
          },
          "T3": {
            "status": "not tested"
          },
          "T4": {
            "status": "not tested"
          },
          "T5": {
            "status": "not tested"
          }
        }
      },
      "public_benchmarks": {
        "lmarena_text": {
          "arena_score": {
            "value": 1516,
            "unit": "Elo-style points",
            "rank": 2,
            "sample_size": 8300,
            "ci_low": 1508,
            "ci_high": 1523,
            "source_entry_name": "Claude Opus 5.5 (High)",
            "url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset/viewer/text/latest",
            "retrieved": "2026-10-17",
            "source_date": "2026-10-15"
          }
        },
        "lmarena_webdev": {
          "arena_score": {
            "value": 1807,
            "unit": "Elo-style points",
            "rank": 1,
            "sample_size": 2775,
            "ci_low": 1793,
            "ci_high": 1821,
            "source_entry_name": "Claude Opus 5.5 (Max)",
            "url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset/viewer/webdev/latest",
            "retrieved": "2026-10-17",
            "source_date": "2026-10-16"
          }
        }
      }
    },
    {
      "id": "claude-sonnet-5-5",
      "name": "Claude Sonnet 5.5",
      "vendor": "Anthropic",
      "license": "Proprietary",
      "access": "closed",
      "own_tests": {
        "status": "not tested",
        "verdict": null,
        "tasks": {
          "T1": {
            "status": "not tested"
          },
          "T2": {
            "status": "not tested"
          },
          "T3": {
            "status": "not tested"
          },
          "T4": {
            "status": "not tested"
          },
          "T5": {
            "status": "not tested"
          }
        }
      },
      "public_benchmarks": {
        "lmarena_text": {
          "arena_score": {
            "value": 1470,
            "unit": "Elo-style points",
            "rank": 30,
            "sample_size": 6831,
            "ci_low": 1463,
            "ci_high": 1478,
            "source_entry_name": "claude-sonnet-5.5-xhigh",
            "url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset/viewer/text/latest",
            "retrieved": "2026-10-17",
            "source_date": "2026-10-15"
          }
        },
        "lmarena_webdev": {
          "arena_score": {
            "value": 1768,
            "unit": "Elo-style points",
            "rank": 3,
            "sample_size": 3182,
            "ci_low": 1754,
            "ci_high": 1781,
            "source_entry_name": "Claude Sonnet 5.5 (xHigh)",
            "url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset/viewer/webdev/latest",
            "retrieved": "2026-10-17",
            "source_date": "2026-10-16"
          }
        }
      }
    },
    {
      "id": "deepseek-v4-pro-0813",
      "name": "DeepSeek V4 Pro (0813)",
      "vendor": "DeepSeek",
      "license": "MIT",
      "access": "open-weight-or-open-license",
      "own_tests": {
        "status": "not tested",
        "verdict": null,
        "tasks": {
          "T1": {
            "status": "not tested"
          },
          "T2": {
            "status": "not tested"
          },
          "T3": {
            "status": "not tested"
          },
          "T4": {
            "status": "not tested"
          },
          "T5": {
            "status": "not tested"
          }
        }
      },
      "public_benchmarks": {
        "lmarena_text": {
          "arena_score": {
            "value": 1446,
            "unit": "Elo-style points",
            "rank": 61,
            "sample_size": 9800,
            "ci_low": 1439,
            "ci_high": 1452,
            "source_entry_name": "deepseek-v4-pro-high-20260813",
            "url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset/viewer/text/latest",
            "retrieved": "2026-10-17",
            "source_date": "2026-10-15"
          }
        },
        "lmarena_webdev": {
          "arena_score": {
            "value": 1582,
            "unit": "Elo-style points",
            "rank": 33,
            "sample_size": 4728,
            "ci_low": 1573,
            "ci_high": 1591,
            "source_entry_name": "DeepSeek-V4-Pro (High, 20260813)",
            "url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset/viewer/webdev/latest",
            "retrieved": "2026-10-17",
            "source_date": "2026-10-16"
          }
        }
      }
    },
    {
      "id": "deepseek-v4-1-flash",
      "name": "DeepSeek V4.1 Flash",
      "vendor": "DeepSeek",
      "license": "MIT",
      "access": "open-weight-or-open-license",
      "own_tests": {
        "status": "not tested",
        "verdict": null,
        "tasks": {
          "T1": {
            "status": "not tested"
          },
          "T2": {
            "status": "not tested"
          },
          "T3": {
            "status": "not tested"
          },
          "T4": {
            "status": "not tested"
          },
          "T5": {
            "status": "not tested"
          }
        }
      },
      "public_benchmarks": {
        "lmarena_text": {
          "arena_score": {
            "value": 1463,
            "unit": "Elo-style points",
            "rank": 38,
            "sample_size": 12793,
            "ci_low": 1457,
            "ci_high": 1469,
            "source_entry_name": "deepseek-v4.1-flash-max",
            "url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset/viewer/text/latest",
            "retrieved": "2026-10-17",
            "source_date": "2026-10-15"
          }
        },
        "lmarena_webdev": {
          "arena_score": {
            "value": 1618,
            "unit": "Elo-style points",
            "rank": 24,
            "sample_size": 3540,
            "ci_low": 1607,
            "ci_high": 1629,
            "source_entry_name": "DeepSeek-V4.1-Flash (Max)",
            "url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset/viewer/webdev/latest",
            "retrieved": "2026-10-17",
            "source_date": "2026-10-16"
          }
        }
      }
    },
    {
      "id": "gemini-3-8-flash",
      "name": "Gemini 3.8 Flash",
      "vendor": "Google",
      "license": "Proprietary",
      "access": "closed",
      "own_tests": {
        "status": "not tested",
        "verdict": null,
        "tasks": {
          "T1": {
            "status": "not tested"
          },
          "T2": {
            "status": "not tested"
          },
          "T3": {
            "status": "not tested"
          },
          "T4": {
            "status": "not tested"
          },
          "T5": {
            "status": "not tested"
          }
        }
      },
      "public_benchmarks": {
        "lmarena_text": {
          "arena_score": {
            "value": 1499,
            "unit": "Elo-style points",
            "rank": 8,
            "sample_size": 37454,
            "ci_low": 1494,
            "ci_high": 1504,
            "source_entry_name": "Gemini 3.8 Flash (High)",
            "url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset/viewer/text/latest",
            "retrieved": "2026-10-17",
            "source_date": "2026-10-15"
          }
        },
        "lmarena_webdev": {
          "arena_score": {
            "value": 1585,
            "unit": "Elo-style points",
            "rank": 31,
            "sample_size": 10514,
            "ci_low": 1578,
            "ci_high": 1593,
            "source_entry_name": "Gemini 3.8 Flash (High)",
            "url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset/viewer/webdev/latest",
            "retrieved": "2026-10-17",
            "source_date": "2026-10-16"
          }
        }
      }
    },
    {
      "id": "gemini-4-argon",
      "name": "Gemini 4 Argon",
      "vendor": "Google",
      "license": "Proprietary",
      "access": "closed",
      "own_tests": {
        "status": "not tested",
        "verdict": null,
        "tasks": {
          "T1": {
            "status": "not tested"
          },
          "T2": {
            "status": "not tested"
          },
          "T3": {
            "status": "not tested"
          },
          "T4": {
            "status": "not tested"
          },
          "T5": {
            "status": "not tested"
          }
        }
      },
      "public_benchmarks": {
        "lmarena_text": {
          "arena_score": {
            "value": 1532,
            "unit": "Elo-style points",
            "rank": 1,
            "sample_size": 4904,
            "ci_low": 1524,
            "ci_high": 1541,
            "source_entry_name": "Gemini 4 Argon (High)",
            "url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset/viewer/text/latest",
            "retrieved": "2026-10-17",
            "source_date": "2026-10-15"
          }
        },
        "lmarena_webdev": {
          "arena_score": {
            "value": 1678,
            "unit": "Elo-style points",
            "rank": 9,
            "sample_size": 1775,
            "ci_low": 1663,
            "ci_high": 1694,
            "source_entry_name": "Gemini 4 Argon (High)",
            "url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset/viewer/webdev/latest",
            "retrieved": "2026-10-17",
            "source_date": "2026-10-16"
          }
        }
      }
    },
    {
      "id": "gemma-4-31b",
      "name": "Gemma 4 31B",
      "vendor": "Google",
      "license": "Apache 2.0",
      "access": "open-weight-or-open-license",
      "own_tests": {
        "status": "not tested",
        "verdict": null,
        "tasks": {
          "T1": {
            "status": "not tested"
          },
          "T2": {
            "status": "not tested"
          },
          "T3": {
            "status": "not tested"
          },
          "T4": {
            "status": "not tested"
          },
          "T5": {
            "status": "not tested"
          }
        }
      },
      "public_benchmarks": {
        "lmarena_text": {
          "arena_score": {
            "value": 1443,
            "unit": "Elo-style points",
            "rank": 67,
            "sample_size": 6147,
            "ci_low": 1435,
            "ci_high": 1450,
            "source_entry_name": "gemma-4-31b",
            "url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset/viewer/text/latest",
            "retrieved": "2026-10-17",
            "source_date": "2026-10-15"
          }
        },
        "lmarena_webdev": {
          "arena_score": {
            "value": 1367,
            "unit": "Elo-style points",
            "rank": 106,
            "sample_size": 14490,
            "ci_low": 1361,
            "ci_high": 1374,
            "source_entry_name": "Gemma 4 31B",
            "url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset/viewer/webdev/latest",
            "retrieved": "2026-10-17",
            "source_date": "2026-10-16"
          }
        }
      }
    },
    {
      "id": "muse-spark-1-3",
      "name": "Muse Spark 1.3",
      "vendor": "Meta",
      "license": "Proprietary",
      "access": "closed",
      "own_tests": {
        "status": "not tested",
        "verdict": null,
        "tasks": {
          "T1": {
            "status": "not tested"
          },
          "T2": {
            "status": "not tested"
          },
          "T3": {
            "status": "not tested"
          },
          "T4": {
            "status": "not tested"
          },
          "T5": {
            "status": "not tested"
          }
        }
      },
      "public_benchmarks": {
        "lmarena_text": {
          "arena_score": {
            "value": 1490,
            "unit": "Elo-style points",
            "rank": 12,
            "sample_size": 17561,
            "ci_low": 1484,
            "ci_high": 1496,
            "source_entry_name": "Muse Spark 1.3 (Max)",
            "url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset/viewer/text/latest",
            "retrieved": "2026-10-17",
            "source_date": "2026-10-15"
          }
        },
        "lmarena_webdev": {
          "arena_score": {
            "value": 1658,
            "unit": "Elo-style points",
            "rank": 12,
            "sample_size": 9068,
            "ci_low": 1650,
            "ci_high": 1666,
            "source_entry_name": "Muse Spark 1.3 (Max)",
            "url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset/viewer/webdev/latest",
            "retrieved": "2026-10-17",
            "source_date": "2026-10-16"
          }
        }
      }
    },
    {
      "id": "kimi-k3",
      "name": "Kimi K3",
      "vendor": "Moonshot AI",
      "license": "Kimi K3 license",
      "access": "open-weight-or-open-license",
      "own_tests": {
        "status": "not tested",
        "verdict": null,
        "tasks": {
          "T1": {
            "status": "not tested"
          },
          "T2": {
            "status": "not tested"
          },
          "T3": {
            "status": "not tested"
          },
          "T4": {
            "status": "not tested"
          },
          "T5": {
            "status": "not tested"
          }
        }
      },
      "public_benchmarks": {
        "lmarena_text": {
          "arena_score": {
            "value": 1475,
            "unit": "Elo-style points",
            "rank": 25,
            "sample_size": 32795,
            "ci_low": 1470,
            "ci_high": 1479,
            "source_entry_name": "Kimi K3 (Max)",
            "url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset/viewer/text/latest",
            "retrieved": "2026-10-17",
            "source_date": "2026-10-15"
          }
        },
        "lmarena_webdev": {
          "arena_score": {
            "value": 1655,
            "unit": "Elo-style points",
            "rank": 14,
            "sample_size": 19220,
            "ci_low": 1649,
            "ci_high": 1661,
            "source_entry_name": "Kimi K3 (Max)",
            "url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset/viewer/webdev/latest",
            "retrieved": "2026-10-17",
            "source_date": "2026-10-16"
          }
        }
      }
    },
    {
      "id": "gpt-6-astra",
      "name": "GPT-6 Astra",
      "vendor": "OpenAI",
      "license": "Proprietary",
      "access": "closed",
      "own_tests": {
        "status": "not tested",
        "verdict": null,
        "tasks": {
          "T1": {
            "status": "not tested"
          },
          "T2": {
            "status": "not tested"
          },
          "T3": {
            "status": "not tested"
          },
          "T4": {
            "status": "not tested"
          },
          "T5": {
            "status": "not tested"
          }
        }
      },
      "public_benchmarks": {
        "lmarena_text": {
          "arena_score": {
            "value": 1443,
            "unit": "Elo-style points",
            "rank": 69,
            "sample_size": 12163,
            "ci_low": 1436,
            "ci_high": 1449,
            "source_entry_name": "gpt-6-astra-max",
            "url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset/viewer/text/latest",
            "retrieved": "2026-10-17",
            "source_date": "2026-10-15"
          }
        },
        "lmarena_webdev": {
          "arena_score": {
            "value": 1777,
            "unit": "Elo-style points",
            "rank": 2,
            "sample_size": 6014,
            "ci_low": 1767,
            "ci_high": 1788,
            "source_entry_name": "GPT-6 Astra (Max)",
            "url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset/viewer/webdev/latest",
            "retrieved": "2026-10-17",
            "source_date": "2026-10-16"
          }
        }
      }
    },
    {
      "id": "gpt-6-luna",
      "name": "GPT-6 Luna",
      "vendor": "OpenAI",
      "license": "Proprietary",
      "access": "closed",
      "own_tests": {
        "status": "not tested",
        "verdict": null,
        "tasks": {
          "T1": {
            "status": "not tested"
          },
          "T2": {
            "status": "not tested"
          },
          "T3": {
            "status": "not tested"
          },
          "T4": {
            "status": "not tested"
          },
          "T5": {
            "status": "not tested"
          }
        }
      },
      "public_benchmarks": {
        "lmarena_text": {
          "arena_score": {
            "value": 1392,
            "unit": "Elo-style points",
            "rank": 166,
            "sample_size": 49617,
            "ci_low": 1387,
            "ci_high": 1396,
            "source_entry_name": "gpt-6-luna-max",
            "url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset/viewer/text/latest",
            "retrieved": "2026-10-17",
            "source_date": "2026-10-15"
          }
        },
        "lmarena_webdev": {
          "arena_score": {
            "value": 1578,
            "unit": "Elo-style points",
            "rank": 35,
            "sample_size": 20045,
            "ci_low": 1572,
            "ci_high": 1584,
            "source_entry_name": "GPT-6 Luna (Max)",
            "url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset/viewer/webdev/latest",
            "retrieved": "2026-10-17",
            "source_date": "2026-10-16"
          }
        }
      }
    },
    {
      "id": "gpt-6-1-sol",
      "name": "GPT-6.1 Sol",
      "vendor": "OpenAI",
      "license": "Proprietary",
      "access": "closed",
      "own_tests": {
        "status": "not tested",
        "verdict": null,
        "tasks": {
          "T1": {
            "status": "not tested"
          },
          "T2": {
            "status": "not tested"
          },
          "T3": {
            "status": "not tested"
          },
          "T4": {
            "status": "not tested"
          },
          "T5": {
            "status": "not tested"
          }
        }
      },
      "public_benchmarks": {
        "lmarena_text": {
          "arena_score": {
            "value": 1448,
            "unit": "Elo-style points",
            "rank": 57,
            "sample_size": 6320,
            "ci_low": 1440,
            "ci_high": 1456,
            "source_entry_name": "gpt-6.1-sol-max",
            "url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset/viewer/text/latest",
            "retrieved": "2026-10-17",
            "source_date": "2026-10-15"
          }
        },
        "lmarena_webdev": {
          "arena_score": {
            "value": 1755,
            "unit": "Elo-style points",
            "rank": 4,
            "sample_size": 2863,
            "ci_low": 1742,
            "ci_high": 1769,
            "source_entry_name": "GPT-6.1 Sol (Max)",
            "url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset/viewer/webdev/latest",
            "retrieved": "2026-10-17",
            "source_date": "2026-10-16"
          }
        }
      }
    },
    {
      "id": "hy4-preview",
      "name": "Hy4 preview",
      "vendor": "Tencent",
      "license": "Apache 2.0",
      "access": "open-weight-or-open-license",
      "own_tests": {
        "status": "not tested",
        "verdict": null,
        "tasks": {
          "T1": {
            "status": "not tested"
          },
          "T2": {
            "status": "not tested"
          },
          "T3": {
            "status": "not tested"
          },
          "T4": {
            "status": "not tested"
          },
          "T5": {
            "status": "not tested"
          }
        }
      },
      "public_benchmarks": {
        "lmarena_webdev": {
          "arena_score": {
            "value": 1631,
            "unit": "Elo-style points",
            "rank": 18,
            "sample_size": 4802,
            "ci_low": 1622,
            "ci_high": 1641,
            "source_entry_name": "Hy4 preview",
            "url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset/viewer/webdev/latest",
            "retrieved": "2026-10-17",
            "source_date": "2026-10-16"
          }
        }
      }
    },
    {
      "id": "grok-4-7",
      "name": "Grok 4.7",
      "vendor": "xAI",
      "license": "Proprietary",
      "access": "closed",
      "own_tests": {
        "status": "not tested",
        "verdict": null,
        "tasks": {
          "T1": {
            "status": "not tested"
          },
          "T2": {
            "status": "not tested"
          },
          "T3": {
            "status": "not tested"
          },
          "T4": {
            "status": "not tested"
          },
          "T5": {
            "status": "not tested"
          }
        }
      },
      "public_benchmarks": {
        "lmarena_text": {
          "arena_score": {
            "value": 1401,
            "unit": "Elo-style points",
            "rank": 153,
            "sample_size": 11591,
            "ci_low": 1394,
            "ci_high": 1407,
            "source_entry_name": "grok-4.7-xhigh",
            "url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset/viewer/text/latest",
            "retrieved": "2026-10-17",
            "source_date": "2026-10-15"
          }
        },
        "lmarena_webdev": {
          "arena_score": {
            "value": 1639,
            "unit": "Elo-style points",
            "rank": 16,
            "sample_size": 5021,
            "ci_low": 1629,
            "ci_high": 1648,
            "source_entry_name": "Grok 4.7 (xHigh)",
            "url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset/viewer/webdev/latest",
            "retrieved": "2026-10-17",
            "source_date": "2026-10-16"
          }
        }
      }
    },
    {
      "id": "mimo-v2-6-flash",
      "name": "MiMo-V2.6 Flash",
      "vendor": "Xiaomi",
      "license": "MIT",
      "access": "open-weight-or-open-license",
      "own_tests": {
        "status": "not tested",
        "verdict": null,
        "tasks": {
          "T1": {
            "status": "not tested"
          },
          "T2": {
            "status": "not tested"
          },
          "T3": {
            "status": "not tested"
          },
          "T4": {
            "status": "not tested"
          },
          "T5": {
            "status": "not tested"
          }
        }
      },
      "public_benchmarks": {
        "lmarena_text": {
          "arena_score": {
            "value": 1455,
            "unit": "Elo-style points",
            "rank": 45,
            "sample_size": 10917,
            "ci_low": 1449,
            "ci_high": 1462,
            "source_entry_name": "mimo-v2.6-flash",
            "url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset/viewer/text/latest",
            "retrieved": "2026-10-17",
            "source_date": "2026-10-15"
          }
        },
        "lmarena_webdev": {
          "arena_score": {
            "value": 1634,
            "unit": "Elo-style points",
            "rank": 17,
            "sample_size": 5016,
            "ci_low": 1625,
            "ci_high": 1644,
            "source_entry_name": "MiMo-V2.6-Flash",
            "url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset/viewer/webdev/latest",
            "retrieved": "2026-10-17",
            "source_date": "2026-10-16"
          }
        }
      }
    },
    {
      "id": "mimo-v2-6-pro",
      "name": "MiMo-V2.6 Pro",
      "vendor": "Xiaomi",
      "license": "MIT",
      "access": "open-weight-or-open-license",
      "own_tests": {
        "status": "not tested",
        "verdict": null,
        "tasks": {
          "T1": {
            "status": "not tested"
          },
          "T2": {
            "status": "not tested"
          },
          "T3": {
            "status": "not tested"
          },
          "T4": {
            "status": "not tested"
          },
          "T5": {
            "status": "not tested"
          }
        }
      },
      "public_benchmarks": {
        "lmarena_text": {
          "arena_score": {
            "value": 1491,
            "unit": "Elo-style points",
            "rank": 11,
            "sample_size": 4075,
            "ci_low": 1481,
            "ci_high": 1500,
            "source_entry_name": "mimo-v2.6-pro",
            "url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset/viewer/text/latest",
            "retrieved": "2026-10-17",
            "source_date": "2026-10-15"
          }
        },
        "lmarena_webdev": {
          "arena_score": {
            "value": 1625,
            "unit": "Elo-style points",
            "rank": 20,
            "sample_size": 4763,
            "ci_low": 1615,
            "ci_high": 1635,
            "source_entry_name": "MiMo-V2.6-Pro",
            "url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset/viewer/webdev/latest",
            "retrieved": "2026-10-17",
            "source_date": "2026-10-16"
          }
        }
      }
    },
    {
      "id": "glm-5-3",
      "name": "GLM-5.3",
      "vendor": "Z.ai",
      "license": "MIT",
      "access": "open-weight-or-open-license",
      "own_tests": {
        "status": "not tested",
        "verdict": null,
        "tasks": {
          "T1": {
            "status": "not tested"
          },
          "T2": {
            "status": "not tested"
          },
          "T3": {
            "status": "not tested"
          },
          "T4": {
            "status": "not tested"
          },
          "T5": {
            "status": "not tested"
          }
        }
      },
      "public_benchmarks": {
        "lmarena_text": {
          "arena_score": {
            "value": 1471,
            "unit": "Elo-style points",
            "rank": 27,
            "sample_size": 22543,
            "ci_low": 1466,
            "ci_high": 1477,
            "source_entry_name": "glm-5.3-max",
            "url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset/viewer/text/latest",
            "retrieved": "2026-10-17",
            "source_date": "2026-10-15"
          }
        },
        "lmarena_webdev": {
          "arena_score": {
            "value": 1621,
            "unit": "Elo-style points",
            "rank": 23,
            "sample_size": 9542,
            "ci_low": 1613,
            "ci_high": 1629,
            "source_entry_name": "GLM-5.3 (Max)",
            "url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset/viewer/webdev/latest",
            "retrieved": "2026-10-17",
            "source_date": "2026-10-16"
          }
        }
      }
    },
    {
      "id": "glm-5-3-flash",
      "name": "GLM-5.3 Flash",
      "vendor": "Z.ai",
      "license": "MIT",
      "access": "open-weight-or-open-license",
      "own_tests": {
        "status": "not tested",
        "verdict": null,
        "tasks": {
          "T1": {
            "status": "not tested"
          },
          "T2": {
            "status": "not tested"
          },
          "T3": {
            "status": "not tested"
          },
          "T4": {
            "status": "not tested"
          },
          "T5": {
            "status": "not tested"
          }
        }
      },
      "public_benchmarks": {
        "lmarena_text": {
          "arena_score": {
            "value": 1471,
            "unit": "Elo-style points",
            "rank": 29,
            "sample_size": 32925,
            "ci_low": 1466,
            "ci_high": 1475,
            "source_entry_name": "glm-5.3-flash",
            "url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset/viewer/text/latest",
            "retrieved": "2026-10-17",
            "source_date": "2026-10-15"
          }
        },
        "lmarena_webdev": {
          "arena_score": {
            "value": 1605,
            "unit": "Elo-style points",
            "rank": 27,
            "sample_size": 17579,
            "ci_low": 1599,
            "ci_high": 1612,
            "source_entry_name": "GLM-5.3-Flash",
            "url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset/viewer/webdev/latest",
            "retrieved": "2026-10-17",
            "source_date": "2026-10-16"
          }
        }
      }
    }
  ]
}