{
  "api_version": "1",
  "updated": "2026-10-29",
  "latest_event": "2026-10-28",
  "rights": "© Frontier Log, all rights reserved.",
  "benchmark": {
    "slug": "livecodebench-v6",
    "name": "LiveCodeBench v6",
    "family": "livecodebench",
    "unit": "%",
    "higher_is_better": true,
    "status": "retired",
    "active": false,
    "superseded_by": null,
    "retired": "2026-09-18",
    "benchmark_url": "https://livecodebench.github.io",
    "introduced": "2025-04-21",
    "creator": "UC Berkeley, MIT and Cornell",
    "introduced_source": "https://huggingface.co/datasets/livecodebench/code_generation_lite/commit/503416f80ac447d31f252d8434023786ce45dd0c",
    "settings": [
      {
        "setting": "default",
        "name": "",
        "url": "https://frontierlog.ai/benchmark/livecodebench-v6",
        "results": 38
      }
    ]
  },
  "total": 38,
  "results": [
    {
      "id": "gemma-3-27b:livecodebench-v6:default:1",
      "model": "gemma-3-27b",
      "model_name": "Gemma 3 27B",
      "model_date": "2025-03-12",
      "event": "https://frontierlog.ai/event/2025-03-12-gemma-3",
      "benchmark": "livecodebench-v6",
      "benchmark_name": "LiveCodeBench v6",
      "setting": "default",
      "setting_name": "",
      "score": 29.1,
      "unit": "%",
      "printed": "29.1%",
      "reported_by": "google",
      "run_by": null,
      "label": "Gemma 3 27B IT",
      "source": "https://blog.google/innovation-and-ai/technology/developers-tools/gemma-4/",
      "source_title": "Gemma 4: Byte for byte, the most capable open models",
      "source_date": "2026-04-02",
      "where": "benchmark table (image), row 'LiveCodeBench v6', column 'Gemma 3 27B IT'",
      "own": false,
      "primary": true,
      "parallel": false,
      "notes": "",
      "caveat": "",
      "human_verified": false,
      "human_verified_on": null,
      "check": "corroborated",
      "corroborated": true,
      "corroborated_on": "2026-10-28",
      "corroborated_by": [
        {
          "url": "https://ai.google.dev/gemma/docs/core/model_card_4",
          "name": "Gemma 4 card",
          "title": "Gemma 4 model card",
          "lab": "google",
          "date": "2026-04-02",
          "printed": "29.1%",
          "where": "Benchmark Results table (instruction-tuned), row 'LiveCodeBench v6', column 'Gemma 3 27B (no think)'"
        },
        {
          "url": "https://huggingface.co/google/gemma-4-12B-it",
          "name": "Gemma 4 12B card",
          "title": "google/gemma-4-12B-it model card",
          "lab": "google",
          "date": "2026-06-03",
          "printed": "29.1%",
          "where": "'Benchmark Results' table (instruction-tuned), row 'LiveCodeBench v6', column 'Gemma 3 27B (no think)'"
        }
      ],
      "source_image": "https://frontierlog.ai/sources/00a53e00c7528f95.jpg",
      "source_image_original": "https://storage.googleapis.com/gweb-uniblog-publish-prod/documents/gemma-4-table_light_Web_with_Arena.jpg"
    },
    {
      "id": "gemini-2-5-pro:livecodebench-v6:default:1",
      "model": "gemini-2-5-pro",
      "model_name": "Gemini 2.5 Pro",
      "model_date": "2025-03-25",
      "event": "https://frontierlog.ai/event/2025-03-25-gemini-2-5-pro",
      "benchmark": "livecodebench-v6",
      "benchmark_name": "LiveCodeBench v6",
      "setting": "default",
      "setting_name": "",
      "score": 74.2,
      "unit": "%",
      "printed": "74.2%",
      "reported_by": "google",
      "run_by": null,
      "label": "Gemini 2.5 Pro",
      "source": "https://storage.googleapis.com/deepmind-media/gemini/gemini_v2_5_report.pdf",
      "source_title": "Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities",
      "source_date": "2025-06-17",
      "where": "Table 3 (core capability benchmarks), row 'Code, LiveCodeBench', column 'Gemini 2.5 Pro'",
      "own": false,
      "primary": true,
      "parallel": false,
      "notes": "pass@1; single attempt; problems from 1/1/2025 to 5/1/2025",
      "caveat": "",
      "human_verified": false,
      "human_verified_on": null,
      "check": "corroborated",
      "corroborated": true,
      "corroborated_on": "2026-10-28",
      "corroborated_by": [
        {
          "url": "https://storage.googleapis.com/deepmind-media/Model-Cards/Gemini-2-5-Deep-Think-Model-Card.pdf",
          "name": "Google brings Deep Think to the Gemini app card",
          "title": "Gemini 2.5 Deep Think Model Card",
          "lab": "google",
          "date": "2025-08-01",
          "printed": "74.2%",
          "where": "Results table, Code generation LiveCodeBench v6 (UI: 1/1/2025-5/1/2025) row, Gemini 2.5 Pro column"
        }
      ]
    },
    {
      "id": "mistral-medium-3:livecodebench-v6:default:1",
      "model": "mistral-medium-3",
      "model_name": "Mistral Medium 3",
      "model_date": "2025-05-07",
      "event": "https://frontierlog.ai/event/2025-05-07-mistral-medium-3",
      "benchmark": "livecodebench-v6",
      "benchmark_name": "LiveCodeBench v6",
      "setting": "default",
      "setting_name": "",
      "score": 30.3,
      "unit": "%",
      "printed": "30.3%",
      "reported_by": "mistral",
      "run_by": null,
      "label": "Mistral Medium 3",
      "source": "https://mistral.ai/news/mistral-medium-3/",
      "source_title": "Medium is the new large.",
      "source_date": "2025-05-07",
      "where": "benchmark table image ('Table Medium 6'), section CODING, row 'LiveCodeBench (v6) 0-shot', column 'Mistral Medium 3'",
      "own": true,
      "primary": true,
      "parallel": false,
      "notes": "0-shot",
      "caveat": "",
      "human_verified": true,
      "human_verified_on": "2026-10-20",
      "check": "human",
      "corroborated": false,
      "corroborated_on": null,
      "corroborated_by": [],
      "source_image": "https://frontierlog.ai/sources/da41e4f28d4c6435.webp",
      "source_image_original": "https://mistral.ai/_astro/1a845a6e-34b8-4a0e-83d8-25a1436241a9_ZfoqEG.webp"
    },
    {
      "id": "magistral-medium:livecodebench-v6:default:1",
      "model": "magistral-medium",
      "model_name": "Magistral Medium",
      "model_date": "2025-06-10",
      "event": "https://frontierlog.ai/event/2025-06-10-magistral",
      "benchmark": "livecodebench-v6",
      "benchmark_name": "LiveCodeBench v6",
      "setting": "default",
      "setting_name": "",
      "score": 50.3,
      "unit": "%",
      "printed": "50.3",
      "reported_by": "mistral",
      "run_by": null,
      "label": "Magistral Medium",
      "source": "https://arxiv.org/abs/2506.10910",
      "source_title": "Magistral",
      "source_date": "2025-06-10",
      "where": "Magistral paper, Table 2 'Results of Magistral Medium trained solely with RL', row 'LiveCodeBench (v6)', column 'Magistral Medium'",
      "own": true,
      "primary": true,
      "parallel": false,
      "notes": "averaged over 16 runs",
      "caveat": "",
      "human_verified": false,
      "human_verified_on": null,
      "check": "double_entry",
      "corroborated": false,
      "corroborated_on": null,
      "corroborated_by": []
    },
    {
      "id": "magistral-small:livecodebench-v6:default:1",
      "model": "magistral-small",
      "model_name": "Magistral Small",
      "model_date": "2025-06-10",
      "event": "https://frontierlog.ai/event/2025-06-10-magistral",
      "benchmark": "livecodebench-v6",
      "benchmark_name": "LiveCodeBench v6",
      "setting": "default",
      "setting_name": "",
      "score": 47.4,
      "unit": "%",
      "printed": "47.4",
      "reported_by": "mistral",
      "run_by": null,
      "label": "SFT + RL (Magistral Small)",
      "source": "https://arxiv.org/abs/2506.10910",
      "source_title": "Magistral",
      "source_date": "2025-06-10",
      "where": "Magistral paper, Table 3, row 'LiveCodeBench (v6)', column 'SFT + RL (Magistral Small)'",
      "own": true,
      "primary": true,
      "parallel": false,
      "notes": "",
      "caveat": "",
      "human_verified": false,
      "human_verified_on": null,
      "check": "double_entry",
      "corroborated": false,
      "corroborated_on": null,
      "corroborated_by": []
    },
    {
      "id": "kimi-k2:livecodebench-v6:default:1",
      "model": "kimi-k2",
      "model_name": "Kimi K2",
      "model_date": "2025-07-11",
      "event": "https://frontierlog.ai/event/2025-07-11-kimi-k2",
      "benchmark": "livecodebench-v6",
      "benchmark_name": "LiveCodeBench v6",
      "setting": "default",
      "setting_name": "",
      "score": 53.7,
      "unit": "%",
      "printed": "53.7",
      "reported_by": "moonshot",
      "run_by": null,
      "label": "Kimi-K2-Instruct",
      "source": "https://www.kimi.ai/blog/kimi-k2",
      "source_title": "Kimi K2: Open Agentic Intelligence",
      "source_date": "2025-07-11",
      "where": "Evaluation Results table (HTML, Kimi-K2-Instruct tab), row 'LiveCodeBench v6 (Aug 24-May 25)', metric 'Pass@1', column 'Kimi-K2-Instruct'",
      "own": true,
      "primary": true,
      "parallel": false,
      "notes": "problems from Aug 2024 to May 2025; non-thinking",
      "caveat": "",
      "human_verified": false,
      "human_verified_on": null,
      "check": "corroborated",
      "corroborated": true,
      "corroborated_on": "2026-10-28",
      "corroborated_by": [
        {
          "url": "https://huggingface.co/moonshotai/Kimi-K2-Instruct",
          "name": "Kimi K2 card",
          "title": "Kimi-K2-Instruct model card",
          "lab": "moonshot",
          "date": "2025-07-11",
          "printed": "53.7",
          "where": "card Instruction model table, column \"Kimi K2 Instruct\", row \"LiveCodeBench v6(Aug 24 - May 25)\""
        },
        {
          "url": "https://github.com/MoonshotAI/Kimi-K2",
          "name": "Kimi K2 card",
          "title": "Kimi K2 GitHub repository",
          "lab": "moonshot",
          "date": "2025-07-11",
          "printed": "53.7",
          "where": "README Instruction model table, column \"Kimi K2 Instruct\", row \"LiveCodeBench v6(Aug 24 - May 25)\""
        },
        {
          "url": "https://github.com/MoonshotAI/Kimi-K2/blob/main/tech_report.pdf",
          "name": "Kimi K2 card",
          "title": "Kimi K2: Open Agentic Intelligence",
          "lab": "moonshot",
          "date": "2025-07-11",
          "printed": "53.7",
          "where": "tech report instruct evaluation table, column \"Kimi-K2-Instruct\", row \"LiveCodeBench v6 (Pass@1)\""
        }
      ],
      "source_capture": "https://frontierlog.ai/captures/77a1bca84d6b4bf2.png"
    },
    {
      "id": "qwen3-235b-a22b-thinking-2507:livecodebench-v6:default:1",
      "model": "qwen3-235b-a22b-thinking-2507",
      "model_name": "Qwen3-235B-A22B-Thinking-2507",
      "model_date": "2025-07-25",
      "event": "https://frontierlog.ai/event/2025-07-25-qwen3-235b-a22b-thinking-2507",
      "benchmark": "livecodebench-v6",
      "benchmark_name": "LiveCodeBench v6",
      "setting": "default",
      "setting_name": "",
      "score": 74.1,
      "unit": "%",
      "printed": "74.1",
      "reported_by": "alibaba",
      "run_by": null,
      "label": "Qwen3-235B-A22B-Thinking-2507",
      "source": "https://www.alizila.com/alibaba-unveils-new-qwen3-models-for-coding-complexing-reasoning-and-machine-translation/",
      "source_title": "Alibaba Unveils New Qwen3 Models for Coding, Complexing Reasoning and Machine Translation",
      "source_date": "2025-07-25",
      "where": "bar chart image in the 'Qwen3-235B' section, group 'LiveCodeBench v6 (25.02-25.05)', first bar value label",
      "own": true,
      "primary": true,
      "parallel": false,
      "notes": "problems from Feb to May 2025",
      "caveat": "",
      "human_verified": false,
      "human_verified_on": null,
      "check": "corroborated",
      "corroborated": true,
      "corroborated_on": "2026-10-26",
      "corroborated_by": [
        {
          "url": "https://huggingface.co/Qwen/Qwen3-235B-A22B-Thinking-2507/blob/4cd68849b2a0fadb84866c703b133aa8c8636130/README.md",
          "name": "Qwen3-235B-A22B-Thinking-2507 card",
          "title": "Qwen/Qwen3-235B-A22B-Thinking-2507 model card (revision of 2025-07-25)",
          "lab": "alibaba",
          "date": "2025-07-25",
          "printed": "74.1",
          "where": "Qwen3-235B-A22B-Thinking-2507 HF model card at revision 4cd68849, Performance table, row 'LiveCodeBench v6 (25.02-25.05)', column 'Qwen3-235B-A22B-Thinking-2507'"
        }
      ],
      "source_image": "https://frontierlog.ai/sources/3ebde273acb680b6.png",
      "source_image_original": "https://alizila.oss-us-west-1.aliyuncs.com/uploads/2025/07/0CA37A90-CD03-4037-AF89-F1548BB1C156.png"
    },
    {
      "id": "glm-4-5:livecodebench-v6:default:1",
      "model": "glm-4-5",
      "model_name": "GLM-4.5",
      "model_date": "2025-07-28",
      "event": "https://frontierlog.ai/event/2025-07-28-glm-4-5",
      "benchmark": "livecodebench-v6",
      "benchmark_name": "LiveCodeBench v6",
      "setting": "default",
      "setting_name": "",
      "score": 63.3,
      "unit": "%",
      "printed": "63.3",
      "reported_by": "zai",
      "run_by": null,
      "label": "GLM-4.5",
      "source": "https://z.ai/blog/glm-4.6",
      "source_title": "GLM-4.6: Advanced Agentic, Reasoning and Coding Capabilities",
      "source_date": "2025-09-30",
      "where": "GLM-4.6 post, benchmark chart image (eight panels), panel 'LiveCodeBench v6', second bar 'GLM-4.5', value label",
      "own": false,
      "primary": true,
      "parallel": false,
      "notes": "128K context",
      "caveat": "",
      "human_verified": true,
      "human_verified_on": "2026-10-19",
      "check": "human",
      "corroborated": false,
      "corroborated_on": null,
      "corroborated_by": [],
      "source_image": "https://frontierlog.ai/sources/e2ec892bfba7273b.png",
      "source_image_original": "https://z-cdn.chatglm.cn/z-blog/glm-4-6/coding_benchmark.png"
    },
    {
      "id": "qwen3-next:livecodebench-v6:default:1",
      "model": "qwen3-next",
      "model_name": "Qwen3-Next",
      "model_date": "2025-09-11",
      "event": "https://frontierlog.ai/event/2025-09-11-qwen3-next",
      "benchmark": "livecodebench-v6",
      "benchmark_name": "LiveCodeBench v6",
      "setting": "default",
      "setting_name": "",
      "score": 68.7,
      "unit": "%",
      "printed": "68.7",
      "reported_by": "alibaba",
      "run_by": null,
      "label": "Qwen3-Next-80B-A3B-Thinking",
      "source": "https://qwen.ai/blog?id=qwen3-next",
      "source_title": "Qwen3-Next: Towards Ultimate Training & Inference Efficiency",
      "source_date": "2025-09-11",
      "where": "'Thinking Model Performance' bar chart image, group 'LiveCodeBench v6 (25.02-25.05)', first bar 'Qwen3-Next-80B-A3B-Thinking' (red), value label",
      "own": true,
      "primary": true,
      "parallel": false,
      "notes": "thinking model; LiveCodeBench v6, problems from Feb to May 2025",
      "caveat": "",
      "human_verified": true,
      "human_verified_on": "2026-10-28",
      "check": "human",
      "corroborated": true,
      "corroborated_on": "2026-10-28",
      "corroborated_by": [
        {
          "url": "https://huggingface.co/Qwen/Qwen3-Next-80B-A3B-Thinking/blob/3e64fd191eaaf4b3e2ccb5519500a0ae515c3e25/README.md",
          "name": "Qwen3-Next card",
          "title": "Qwen/Qwen3-Next-80B-A3B-Thinking model card (launch-day revision)",
          "lab": "alibaba",
          "date": "2025-09-11",
          "printed": "68.7",
          "where": "eval table, row LiveCodeBench v6 (25.02-25.05), column Qwen3-Next-80B-A3B-Thinking"
        }
      ],
      "source_image": "https://frontierlog.ai/sources/1d75ce113070f7a6.jpeg",
      "source_image_original": "https://qianwen-res.oss-accelerate.aliyuncs.com/Qwen3-Next/thinking_performance.jpeg"
    },
    {
      "id": "qwen3-next:livecodebench-v6:default:2",
      "model": "qwen3-next",
      "model_name": "Qwen3-Next",
      "model_date": "2025-09-11",
      "event": "https://frontierlog.ai/event/2025-09-11-qwen3-next",
      "benchmark": "livecodebench-v6",
      "benchmark_name": "LiveCodeBench v6",
      "setting": "default",
      "setting_name": "",
      "score": 56.6,
      "unit": "%",
      "printed": "56.6",
      "reported_by": "alibaba",
      "run_by": null,
      "label": "Qwen3-Next-80B-A3B-Instruct",
      "source": "https://qwen.ai/blog?id=qwen3-next",
      "source_title": "Qwen3-Next: Towards Ultimate Training & Inference Efficiency",
      "source_date": "2025-09-11",
      "where": "'Instruct Model Performance' bar chart image, group 'LiveCodeBench v6 (25.02-25.05)', first bar 'Qwen3-Next-80B-A3B-Instruct' (red), value label",
      "own": true,
      "primary": false,
      "parallel": false,
      "notes": "instruct model; LiveCodeBench v6, problems from Feb to May 2025",
      "caveat": "",
      "human_verified": true,
      "human_verified_on": "2026-10-28",
      "check": "human",
      "corroborated": false,
      "corroborated_on": null,
      "corroborated_by": [],
      "source_image": "https://frontierlog.ai/sources/d278ea627323bd1c.jpeg",
      "source_image_original": "https://qianwen-res.oss-accelerate.aliyuncs.com/Qwen3-Next/Qwen3-Next-80B-A3B-Instruct.001.jpeg"
    },
    {
      "id": "qwen3-vl-235b-a22b-thinking:livecodebench-v6:default:1",
      "model": "qwen3-vl-235b-a22b-thinking",
      "model_name": "Qwen3-VL-235B-A22B-Thinking",
      "model_date": "2025-09-23",
      "event": "https://frontierlog.ai/event/2025-09-23-qwen3-vl",
      "benchmark": "livecodebench-v6",
      "benchmark_name": "LiveCodeBench v6",
      "setting": "default",
      "setting_name": "",
      "score": 70.1,
      "unit": "%",
      "printed": "70.1",
      "reported_by": "alibaba",
      "run_by": null,
      "label": "Qwen3-VL 235B-A22B Thinking",
      "source": "https://qwen.ai/blog?id=qwen3-vl",
      "source_title": "Qwen3-VL: Sharper Vision, Deeper Thought, Broader Action",
      "source_date": "2025-09-23",
      "where": "thinking text-benchmark table image (fourth table, after 'On text-centric tasks'), Reasoning block, row 'LCB v6', column 'Qwen3-VL 235B-A22B Thinking'",
      "own": true,
      "primary": true,
      "parallel": false,
      "notes": "",
      "caveat": "",
      "human_verified": true,
      "human_verified_on": "2026-10-28",
      "check": "human",
      "corroborated": false,
      "corroborated_on": null,
      "corroborated_by": [],
      "source_image": "https://frontierlog.ai/sources/4cce76a4c25eaef1.jpg",
      "source_image_original": "https://qianwen-res.oss-accelerate.aliyuncs.com/Qwen3-VL/table_thinking_text.jpg"
    },
    {
      "id": "qwen3-max:livecodebench-v6:default:1",
      "model": "qwen3-max",
      "model_name": "Qwen3-Max",
      "model_date": "2025-09-24",
      "event": "https://frontierlog.ai/event/2025-09-24-qwen3-max",
      "benchmark": "livecodebench-v6",
      "benchmark_name": "LiveCodeBench v6",
      "setting": "default",
      "setting_name": "",
      "score": 69.0,
      "unit": "%",
      "printed": "69.0",
      "reported_by": "alibaba",
      "run_by": null,
      "label": "Qwen3-Max-Instruct",
      "source": "https://qwen.ai/blog?id=qwen3-max",
      "source_title": "Qwen3-Max: Just Scale it",
      "source_date": "2025-09-24",
      "where": "'Qwen3-Max-Instruct' section, bar chart image, group 'LiveCodeBench v6 (25.02-25.05)', first bar 'Qwen3-Max-Instruct' (red), value label",
      "own": true,
      "primary": true,
      "parallel": false,
      "notes": "problems from Feb to May 2025",
      "caveat": "",
      "human_verified": true,
      "human_verified_on": "2026-10-19",
      "check": "human",
      "corroborated": false,
      "corroborated_on": null,
      "corroborated_by": [],
      "source_image": "https://frontierlog.ai/sources/2217e1ccbb128cc8.png",
      "source_image_original": "https://qianwen-res.oss-accelerate.aliyuncs.com/Qwen3-Max/Qwen3-Max-Instruct.png"
    },
    {
      "id": "glm-4-6:livecodebench-v6:default:1",
      "model": "glm-4-6",
      "model_name": "GLM-4.6",
      "model_date": "2025-09-30",
      "event": "https://frontierlog.ai/event/2025-09-30-glm-4-6",
      "benchmark": "livecodebench-v6",
      "benchmark_name": "LiveCodeBench v6",
      "setting": "default",
      "setting_name": "",
      "score": 82.8,
      "unit": "%",
      "printed": "82.8",
      "reported_by": "zai",
      "run_by": null,
      "label": "GLM-4.6",
      "source": "https://z.ai/blog/glm-4.6",
      "source_title": "GLM-4.6: Advanced Agentic, Reasoning and Coding Capabilities",
      "source_date": "2025-09-30",
      "where": "benchmark chart image (eight panels), panel 'LiveCodeBench v6', first bar 'GLM-4.6', value label inside the bar",
      "own": true,
      "primary": true,
      "parallel": false,
      "notes": "no tools; 128K context",
      "caveat": "",
      "human_verified": true,
      "human_verified_on": "2026-10-19",
      "check": "human",
      "corroborated": true,
      "corroborated_on": "2026-10-19",
      "corroborated_by": [
        {
          "url": "https://z.ai/blog/glm-4.7",
          "name": "GLM-4.7",
          "title": "GLM-4.7: Advancing the Coding Capability",
          "lab": "zai",
          "date": "2025-12-22",
          "printed": "82.8",
          "where": "Benchmark table, row 'LiveCodeBench-v6', column 'GLM-4.6'"
        }
      ],
      "source_image": "https://frontierlog.ai/sources/e2ec892bfba7273b.png",
      "source_image_original": "https://z-cdn.chatglm.cn/z-blog/glm-4-6/coding_benchmark.png"
    },
    {
      "id": "glm-4-6:livecodebench-v6:default:2",
      "model": "glm-4-6",
      "model_name": "GLM-4.6",
      "model_date": "2025-09-30",
      "event": "https://frontierlog.ai/event/2025-09-30-glm-4-6",
      "benchmark": "livecodebench-v6",
      "benchmark_name": "LiveCodeBench v6",
      "setting": "default",
      "setting_name": "",
      "score": 84.5,
      "unit": "%",
      "printed": "84.5",
      "reported_by": "zai",
      "run_by": null,
      "label": "GLM-4.6",
      "source": "https://z.ai/blog/glm-4.6",
      "source_title": "GLM-4.6: Advanced Agentic, Reasoning and Coding Capabilities",
      "source_date": "2025-09-30",
      "where": "benchmark bar chart (image), group 'LiveCodeBench v6', upper value label 'w/ Tools' on the GLM-4.6 bar",
      "own": true,
      "primary": false,
      "parallel": false,
      "notes": "with tools; 128K context",
      "caveat": "",
      "human_verified": false,
      "human_verified_on": null,
      "check": "corroborated",
      "corroborated": true,
      "corroborated_on": "2026-10-28",
      "corroborated_by": [
        {
          "url": "https://docs.z.ai/guides/llm/glm-4.6",
          "name": "GLM-4.6 card",
          "title": "GLM-4.6 developer docs",
          "lab": "zai",
          "date": "2025-09-30",
          "printed": "84.5",
          "where": "Benchmark chart image (8 benchmarks, 128K context), LiveCodeBench v6 panel, GLM-4.6 bar, upper value labelled 'w/ Tools' (base bar 82.8 is without tools; the GLM-4.7 card's table prints that 82.8 for GLM-4.6 LiveCodeBench-v6)"
        }
      ],
      "source_image": "https://frontierlog.ai/sources/e2ec892bfba7273b.png",
      "source_image_original": "https://z-cdn.chatglm.cn/z-blog/glm-4-6/coding_benchmark.png"
    },
    {
      "id": "kimi-k2-thinking:livecodebench-v6:default:1",
      "model": "kimi-k2-thinking",
      "model_name": "Kimi K2 Thinking",
      "model_date": "2025-11-06",
      "event": "https://frontierlog.ai/event/2025-11-06-kimi-k2-thinking",
      "benchmark": "livecodebench-v6",
      "benchmark_name": "LiveCodeBench v6",
      "setting": "default",
      "setting_name": "",
      "score": 83.1,
      "unit": "%",
      "printed": "83.1",
      "reported_by": "moonshot",
      "run_by": null,
      "label": "K2 Thinking",
      "source": "https://www.kimi.ai/blog/kimi-k2-thinking",
      "source_title": "Introducing Kimi K2 Thinking",
      "source_date": "2025-11-06",
      "where": "Full Evaluations table (HTML), row 'LiveCodeBench v6', sub-row 'no tools', column 'K2 Thinking'",
      "own": true,
      "primary": true,
      "parallel": false,
      "notes": "128K thinking budget; INT4",
      "caveat": "",
      "human_verified": false,
      "human_verified_on": null,
      "check": "double_entry",
      "corroborated": false,
      "corroborated_on": null,
      "corroborated_by": [],
      "source_capture": "https://frontierlog.ai/captures/eff473ae7ab6e3a4.png"
    },
    {
      "id": "nemotron-3-nano-30b-a3b:livecodebench-v6:default:1",
      "model": "nemotron-3-nano-30b-a3b",
      "model_name": "Nemotron 3 Nano 30B-A3B",
      "model_date": "2025-12-15",
      "event": "https://frontierlog.ai/event/2025-12-15-nemotron-3",
      "benchmark": "livecodebench-v6",
      "benchmark_name": "LiveCodeBench v6",
      "setting": "default",
      "setting_name": "",
      "score": 68.3,
      "unit": "%",
      "printed": "68.3",
      "reported_by": "nvidia",
      "run_by": null,
      "label": "NVIDIA-Nemotron-3-Nano-30B-A3B-BF16",
      "source": "https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-FP8/blob/fa5699ff0bb9da723217af52488c0f9653512ffb/README.md",
      "source_title": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-FP8 model card (launch-day revision)",
      "source_date": "2025-12-15",
      "where": "'Reasoning Benchmark Evaluations' table, row 'LiveCodeBench (v6 ...)', column 'NVIDIA-Nemotron-3-Nano-30B-A3B-BF16'",
      "own": true,
      "primary": true,
      "parallel": false,
      "notes": "",
      "caveat": "",
      "human_verified": false,
      "human_verified_on": null,
      "check": "corroborated",
      "corroborated": true,
      "corroborated_on": "2026-10-19",
      "corroborated_by": [
        {
          "url": "https://research.nvidia.com/labs/nemotron/files/NVIDIA-Nemotron-3-Nano-Technical-Report.pdf",
          "name": "Nemotron 3 card",
          "title": "Nemotron 3 Nano: Open, Efficient Mixture-of-Experts Hybrid Mamba-Transformer Model for Agentic Reasoning",
          "lab": "nvidia",
          "date": "2025-12-15",
          "printed": "68.25",
          "where": "Table 3, LiveCodeBench (v6 2024-08↔2025-05) row, N-3-Nano column"
        },
        {
          "url": "https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16/blob/bf77c3174f68ad409e1c2aa60daeb46e32d1c606/README.md",
          "name": "Nemotron 3 card",
          "title": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16 model card (revision of 2026-08-24)",
          "lab": "nvidia",
          "date": "2025-12-15",
          "printed": "68.3",
          "where": "Benchmarks table, 'Nemotron-3-Nano' (first, BF16) column, LiveCodeBench (v6 2025-08–2025-05) row"
        }
      ],
      "source_capture": "https://frontierlog.ai/captures/4ff9ee8b68e61bc2.png"
    },
    {
      "id": "nemotron-3-nano-30b-a3b:livecodebench-v6:default:2",
      "model": "nemotron-3-nano-30b-a3b",
      "model_name": "Nemotron 3 Nano 30B-A3B",
      "model_date": "2025-12-15",
      "event": "https://frontierlog.ai/event/2025-12-15-nemotron-3",
      "benchmark": "livecodebench-v6",
      "benchmark_name": "LiveCodeBench v6",
      "setting": "default",
      "setting_name": "",
      "score": 68.2,
      "unit": "%",
      "printed": "68.2",
      "reported_by": "nvidia",
      "run_by": null,
      "label": "Nemotron-3-Nano-30B-A3B",
      "source": "https://research.nvidia.com/labs/nemotron/Nemotron-3/",
      "source_title": "NVIDIA Nemotron 3 Family of Models",
      "source_date": "2025-12-15",
      "where": "comparison bar chart (image) under the Nemotron 3 Nano section, group 'LCB v6 (Coding)', value label of the green 'Nemotron-3-Nano-30B-A3B' bar",
      "own": true,
      "primary": false,
      "parallel": false,
      "notes": "",
      "caveat": "",
      "human_verified": true,
      "human_verified_on": "2026-10-20",
      "check": "human",
      "corroborated": false,
      "corroborated_on": null,
      "corroborated_by": [],
      "source_image": "https://frontierlog.ai/sources/e1cd447601f788eb.png",
      "source_image_original": "https://research.nvidia.com/labs/nemotron/images/nvidia-nemotron-3-nano/nano-3-comparison.png"
    },
    {
      "id": "glm-4-7:livecodebench-v6:default:1",
      "model": "glm-4-7",
      "model_name": "GLM-4.7",
      "model_date": "2025-12-22",
      "event": "https://frontierlog.ai/event/2025-12-22-glm-4-7",
      "benchmark": "livecodebench-v6",
      "benchmark_name": "LiveCodeBench v6",
      "setting": "default",
      "setting_name": "",
      "score": 84.9,
      "unit": "%",
      "printed": "84.9",
      "reported_by": "zai",
      "run_by": null,
      "label": "GLM-4.7",
      "source": "https://z.ai/blog/glm-4.7",
      "source_title": "GLM-4.7: Advancing the Coding Capability",
      "source_date": "2025-12-22",
      "where": "benchmark table, row 'LiveCodeBench-v6', column 'GLM-4.7'",
      "own": true,
      "primary": true,
      "parallel": false,
      "notes": "",
      "caveat": "",
      "human_verified": true,
      "human_verified_on": "2026-10-19",
      "check": "human",
      "corroborated": false,
      "corroborated_on": null,
      "corroborated_by": [],
      "source_capture": "https://frontierlog.ai/captures/b52f90e840f8cbd6.png"
    },
    {
      "id": "qwen3-max-thinking:livecodebench-v6:default:1",
      "model": "qwen3-max-thinking",
      "model_name": "Qwen3-Max-Thinking",
      "model_date": "2026-01-26",
      "event": "https://frontierlog.ai/event/2026-01-26-qwen3-max-thinking",
      "benchmark": "livecodebench-v6",
      "benchmark_name": "LiveCodeBench v6",
      "setting": "default",
      "setting_name": "",
      "score": 85.9,
      "unit": "%",
      "printed": "85.9",
      "reported_by": "alibaba",
      "run_by": null,
      "label": "Qwen3-Max-Thinking",
      "source": "https://qwen.ai/blog?id=qwen3-max-thinking",
      "source_title": "Pushing Qwen3-Max-Thinking Beyond its Limits",
      "source_date": "2026-01-26",
      "where": "evaluation table, row 'LiveCodeBench v6', column 'Qwen3-Max-Thinking'",
      "own": true,
      "primary": true,
      "parallel": false,
      "notes": "problems from Feb to May 2025",
      "caveat": "",
      "human_verified": true,
      "human_verified_on": "2026-10-19",
      "check": "human",
      "corroborated": false,
      "corroborated_on": null,
      "corroborated_by": [],
      "source_capture": "https://frontierlog.ai/captures/aed8db26403e0575.png"
    },
    {
      "id": "qwen3-max-thinking:livecodebench-v6:default:2",
      "model": "qwen3-max-thinking",
      "model_name": "Qwen3-Max-Thinking",
      "model_date": "2026-01-26",
      "event": "https://frontierlog.ai/event/2026-01-26-qwen3-max-thinking",
      "benchmark": "livecodebench-v6",
      "benchmark_name": "LiveCodeBench v6",
      "setting": "default",
      "setting_name": "",
      "score": 91.4,
      "unit": "%",
      "printed": "91.4",
      "reported_by": "alibaba",
      "run_by": null,
      "label": "Qwen3-Max-Thinking (with TTS)",
      "source": "https://qwen.ai/blog?id=qwen3-max-thinking",
      "source_title": "Pushing Qwen3-Max-Thinking Beyond its Limits",
      "source_date": "2026-01-26",
      "where": "score chart image (seven panels, below the banner), panel 'LiveCodeBench (2025.02-2025.05)', first bar, upper value label (legend 'Qwen3-Max-Thinking (with TTS)')",
      "own": true,
      "primary": false,
      "parallel": true,
      "notes": "heavy mode: test-time scaling over parallel trajectories",
      "caveat": "",
      "human_verified": true,
      "human_verified_on": "2026-10-19",
      "check": "human",
      "corroborated": false,
      "corroborated_on": null,
      "corroborated_by": [],
      "source_image": "https://frontierlog.ai/sources/80d8b6e334ec18d8.png",
      "source_image_original": "https://qianwen-res.oss-accelerate-overseas.aliyuncs.com/Qwen3-Max-Thinking/score.png"
    },
    {
      "id": "qwen3-max-thinking:livecodebench-v6:default:3",
      "model": "qwen3-max-thinking",
      "model_name": "Qwen3-Max-Thinking",
      "model_date": "2026-01-26",
      "event": "https://frontierlog.ai/event/2026-01-26-qwen3-max-thinking",
      "benchmark": "livecodebench-v6",
      "benchmark_name": "LiveCodeBench v6",
      "setting": "default",
      "setting_name": "",
      "score": 88.0,
      "unit": "%",
      "printed": "88.0",
      "reported_by": "alibaba",
      "run_by": null,
      "label": "Qwen3-Max-Thinking",
      "source": "https://qwen.ai/blog?id=qwen3-max-thinking",
      "source_title": "Pushing Qwen3-Max-Thinking Beyond its Limits",
      "source_date": "2026-01-26",
      "where": "'Test-time Scaling Strategy' section, last sentence of the paragraph, first number in 'LiveCodeBench v6 (88.0 → 91.4)'",
      "own": true,
      "primary": false,
      "parallel": true,
      "notes": "standard parallel sampling and aggregation, the baseline for heavy mode",
      "caveat": "",
      "human_verified": true,
      "human_verified_on": "2026-10-28",
      "check": "human",
      "corroborated": false,
      "corroborated_on": null,
      "corroborated_by": []
    },
    {
      "id": "kimi-k2-5:livecodebench-v6:default:1",
      "model": "kimi-k2-5",
      "model_name": "Kimi K2.5",
      "model_date": "2026-01-27",
      "event": "https://frontierlog.ai/event/2026-01-27-kimi-k2-5",
      "benchmark": "livecodebench-v6",
      "benchmark_name": "LiveCodeBench v6",
      "setting": "default",
      "setting_name": "",
      "score": 85.0,
      "unit": "%",
      "printed": "85.0",
      "reported_by": "moonshot",
      "run_by": null,
      "label": "Kimi K2.5 (Thinking)",
      "source": "https://www.kimi.ai/blog/kimi-k2-5",
      "source_title": "Kimi K2.5: Visual Agentic Intelligence",
      "source_date": "2026-01-27",
      "where": "Benchmark table (HTML), row 'LiveCodeBench (v6)', column 'Kimi K2.5 (Thinking)'",
      "own": true,
      "primary": true,
      "parallel": false,
      "notes": "thinking mode; average of 5 runs",
      "caveat": "",
      "human_verified": false,
      "human_verified_on": null,
      "check": "double_entry",
      "corroborated": false,
      "corroborated_on": null,
      "corroborated_by": [],
      "source_capture": "https://frontierlog.ai/captures/982a5c772bc0060d.png"
    },
    {
      "id": "qwen3-5-397b-a17b:livecodebench-v6:default:1",
      "model": "qwen3-5-397b-a17b",
      "model_name": "Qwen3.5-397B-A17B",
      "model_date": "2026-02-16",
      "event": "https://frontierlog.ai/event/2026-02-16-qwen3-5",
      "benchmark": "livecodebench-v6",
      "benchmark_name": "LiveCodeBench v6",
      "setting": "default",
      "setting_name": "",
      "score": 83.6,
      "unit": "%",
      "printed": "83.6",
      "reported_by": "alibaba",
      "run_by": null,
      "label": "Qwen3.5-397B-A17B",
      "source": "https://qwen.ai/blog?id=qwen3.5",
      "source_title": "Qwen3.5: Towards Native Multimodal Agents",
      "source_date": "2026-02-16",
      "where": "'Performance' section, 'Language' table, row 'LiveCodeBench v6', column 'Qwen3.5-397B-A17B'",
      "own": true,
      "primary": true,
      "parallel": false,
      "notes": "",
      "caveat": "",
      "human_verified": true,
      "human_verified_on": "2026-10-19",
      "check": "human",
      "corroborated": false,
      "corroborated_on": null,
      "corroborated_by": [],
      "source_capture": "https://frontierlog.ai/captures/a2e98da83152d308.png"
    },
    {
      "id": "nemotron-3-super-120b-a12b:livecodebench-v6:default:1",
      "model": "nemotron-3-super-120b-a12b",
      "model_name": "Nemotron 3 Super 120B-A12B",
      "model_date": "2026-03-11",
      "event": "https://frontierlog.ai/event/2026-03-11-nemotron-3-super",
      "benchmark": "livecodebench-v6",
      "benchmark_name": "LiveCodeBench v6",
      "setting": "default",
      "setting_name": "",
      "score": 78.69,
      "unit": "%",
      "printed": "78.69",
      "reported_by": "nvidia",
      "run_by": null,
      "label": "Nemotron-3-Super",
      "source": "https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-FP8/blob/1408abd63ba6ba1c37a78b94d332dac35da45370/README.md",
      "source_title": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-FP8 model card (launch-day revision)",
      "source_date": "2026-03-11",
      "where": "benchmark table under the accuracy chart, row 'LiveCodeBench (v6 2024-08...2025-05)', column 'Nemotron-3-Super'",
      "own": true,
      "primary": true,
      "parallel": false,
      "notes": "problems from Aug 2024 to May 2025",
      "caveat": "",
      "human_verified": false,
      "human_verified_on": null,
      "check": "corroborated",
      "corroborated": true,
      "corroborated_on": "2026-10-19",
      "corroborated_by": [
        {
          "url": "https://research.nvidia.com/labs/nemotron/files/NVIDIA-Nemotron-3-Super-Technical-Report.pdf",
          "name": "Nemotron 3 Super card",
          "title": "Nemotron 3 Super: Open, Efficient Mixture-of-Experts Hybrid Mamba-Transformer Model for Agentic Reasoning",
          "lab": "nvidia",
          "date": "2026-03-11",
          "printed": "78.69",
          "where": "Table 8, LiveCodeBench (v6 2024-08↔2025-05) row, N-3-Super column"
        },
        {
          "url": "https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4/blob/ff433f5493e25d631c9f12b5d55c674229923d02/README.md",
          "name": "Nemotron 3 Super card",
          "title": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4 model card (revision of 2026-08-24)",
          "lab": "nvidia",
          "date": "2026-03-11",
          "printed": "78.69",
          "where": "accuracy table, 'Nemotron-3-Super' (BF16) column, LiveCodeBench (v6 2024-08↔2025-05) row"
        }
      ],
      "source_capture": "https://frontierlog.ai/captures/0a1e755f970190a7.png"
    },
    {
      "id": "qwen3-5-omni:livecodebench-v6:default:1",
      "model": "qwen3-5-omni",
      "model_name": "Qwen3.5-Omni",
      "model_date": "2026-03-30",
      "event": "https://frontierlog.ai/event/2026-03-30-qwen3-5-omni",
      "benchmark": "livecodebench-v6",
      "benchmark_name": "LiveCodeBench v6",
      "setting": "default",
      "setting_name": "",
      "score": 65.6,
      "unit": "%",
      "printed": "65.6",
      "reported_by": "alibaba",
      "run_by": null,
      "label": "Qwen3.5-Omni-Plus",
      "source": "https://qwen.ai/blog?id=qwen3.5-omni",
      "source_title": "Qwen3.5-Omni: Scaling Up, Toward Native Omni-Modal AGI",
      "source_date": "2026-03-30",
      "where": "'Text' table (HTML), Reasoning block, row 'LiveCodeBench v6', column 'Qwen3.5-Omni-Plus'",
      "own": true,
      "primary": true,
      "parallel": false,
      "notes": "Qwen3.5-Omni-Plus; no thinking",
      "caveat": "",
      "human_verified": true,
      "human_verified_on": "2026-10-26",
      "check": "human",
      "corroborated": false,
      "corroborated_on": null,
      "corroborated_by": [],
      "source_capture": "https://frontierlog.ai/captures/f1a71324f95519d3.png"
    },
    {
      "id": "qwen3-5-omni-flash:livecodebench-v6:default:1",
      "model": "qwen3-5-omni-flash",
      "model_name": "Qwen3.5-Omni-Flash",
      "model_date": "2026-03-30",
      "event": "https://frontierlog.ai/event/2026-03-30-qwen3-5-omni",
      "benchmark": "livecodebench-v6",
      "benchmark_name": "LiveCodeBench v6",
      "setting": "default",
      "setting_name": "",
      "score": 56.6,
      "unit": "%",
      "printed": "56.6",
      "reported_by": "alibaba",
      "run_by": null,
      "label": "Qwen3.5-Omni-Flash",
      "source": "https://qwen.ai/blog?id=qwen3.5-omni",
      "source_title": "Qwen3.5-Omni: Scaling Up, Toward Native Omni-Modal AGI",
      "source_date": "2026-03-30",
      "where": "'Text' table (HTML), Reasoning block, row 'LiveCodeBench v6', column 'Qwen3.5-Omni-Flash'",
      "own": true,
      "primary": true,
      "parallel": false,
      "notes": "Qwen3.5-Omni-Flash; no thinking",
      "caveat": "",
      "human_verified": true,
      "human_verified_on": "2026-10-28",
      "check": "human",
      "corroborated": false,
      "corroborated_on": null,
      "corroborated_by": [],
      "source_capture": "https://frontierlog.ai/captures/f1a71324f95519d3.png"
    },
    {
      "id": "gemma-4-31b:livecodebench-v6:default:1",
      "model": "gemma-4-31b",
      "model_name": "Gemma 4 31B",
      "model_date": "2026-04-02",
      "event": "https://frontierlog.ai/event/2026-04-02-gemma-4",
      "benchmark": "livecodebench-v6",
      "benchmark_name": "LiveCodeBench v6",
      "setting": "default",
      "setting_name": "",
      "score": 80.0,
      "unit": "%",
      "printed": "80.0%",
      "reported_by": "google",
      "run_by": null,
      "label": "Gemma 4 31B IT",
      "source": "https://blog.google/innovation-and-ai/technology/developers-tools/gemma-4/",
      "source_title": "Gemma 4: Byte for byte, the most capable open models",
      "source_date": "2026-04-02",
      "where": "benchmark table (image), row 'LiveCodeBench v6', column 'Gemma 4 31B IT Thinking'",
      "own": true,
      "primary": true,
      "parallel": false,
      "notes": "thinking",
      "caveat": "",
      "human_verified": false,
      "human_verified_on": null,
      "check": "corroborated",
      "corroborated": true,
      "corroborated_on": "2026-10-28",
      "corroborated_by": [
        {
          "url": "https://ai.google.dev/gemma/docs/core/model_card_4",
          "name": "Gemma 4 card",
          "title": "Gemma 4 model card",
          "lab": "google",
          "date": "2026-04-02",
          "printed": "80.0%",
          "where": "Benchmark Results table (instruction-tuned), row 'LiveCodeBench v6', column 'Gemma 4 31B'"
        },
        {
          "url": "https://huggingface.co/google/gemma-4-12B-it",
          "name": "Gemma 4 12B card",
          "title": "google/gemma-4-12B-it model card",
          "lab": "google",
          "date": "2026-06-03",
          "printed": "80.0%",
          "where": "'Benchmark Results' table (instruction-tuned), row 'LiveCodeBench v6', column 'Gemma 4 31B'"
        }
      ],
      "source_image": "https://frontierlog.ai/sources/00a53e00c7528f95.jpg",
      "source_image_original": "https://storage.googleapis.com/gweb-uniblog-publish-prod/documents/gemma-4-table_light_Web_with_Arena.jpg"
    },
    {
      "id": "qwen3-6-plus:livecodebench-v6:default:1",
      "model": "qwen3-6-plus",
      "model_name": "Qwen3.6-Plus",
      "model_date": "2026-04-02",
      "event": "https://frontierlog.ai/event/2026-04-02-qwen3-6-plus",
      "benchmark": "livecodebench-v6",
      "benchmark_name": "LiveCodeBench v6",
      "setting": "default",
      "setting_name": "",
      "score": 87.1,
      "unit": "%",
      "printed": "87.1",
      "reported_by": "alibaba",
      "run_by": null,
      "label": "Qwen3.6-Plus",
      "source": "https://qwen.ai/blog?id=qwen3.6",
      "source_title": "Qwen3.6-Plus: Towards Real World Agents",
      "source_date": "2026-04-02",
      "where": "'Performance' section, 'Language' table, row 'LiveCodeBench v6', column 'Qwen3.6-Plus'",
      "own": true,
      "primary": true,
      "parallel": false,
      "notes": "",
      "caveat": "",
      "human_verified": true,
      "human_verified_on": "2026-10-19",
      "check": "human",
      "corroborated": false,
      "corroborated_on": null,
      "corroborated_by": [],
      "source_capture": "https://frontierlog.ai/captures/6f4855b376c670b4.png"
    },
    {
      "id": "qwen3-6-35b-a3b:livecodebench-v6:default:1",
      "model": "qwen3-6-35b-a3b",
      "model_name": "Qwen3.6-35B-A3B",
      "model_date": "2026-04-15",
      "event": "https://frontierlog.ai/event/2026-04-15-qwen3-6-35b-a3b",
      "benchmark": "livecodebench-v6",
      "benchmark_name": "LiveCodeBench v6",
      "setting": "default",
      "setting_name": "",
      "score": 80.4,
      "unit": "%",
      "printed": "80.4",
      "reported_by": "alibaba",
      "run_by": null,
      "label": "Qwen3.6-35BA3B",
      "source": "https://qwen.ai/blog?id=qwen3.6-35b-a3b",
      "source_title": "Qwen3.6-35B-A3B: Agentic Coding Power, Now Open to All",
      "source_date": "2026-04-15",
      "where": "'Language' table (HTML), STEM & Reasoning block, row 'LiveCodeBench v6', column 'Qwen3.6-35BA3B'",
      "own": true,
      "primary": true,
      "parallel": false,
      "notes": "",
      "caveat": "",
      "human_verified": true,
      "human_verified_on": "2026-10-28",
      "check": "human",
      "corroborated": true,
      "corroborated_on": "2026-10-28",
      "corroborated_by": [
        {
          "url": "https://huggingface.co/Qwen/Qwen3.6-35B-A3B",
          "name": "Qwen3.6-35B-A3B card",
          "title": "Qwen/Qwen3.6-35B-A3B model card",
          "lab": "alibaba",
          "date": "2026-04-15",
          "printed": "80.4",
          "where": "first eval table, row LiveCodeBench v6, column Qwen3.6-35BA3B"
        }
      ],
      "source_capture": "https://frontierlog.ai/captures/a1cd0de07b87140b.png"
    },
    {
      "id": "kimi-k2-6:livecodebench-v6:default:1",
      "model": "kimi-k2-6",
      "model_name": "Kimi K2.6",
      "model_date": "2026-04-20",
      "event": "https://frontierlog.ai/event/2026-04-20-kimi-k2-6",
      "benchmark": "livecodebench-v6",
      "benchmark_name": "LiveCodeBench v6",
      "setting": "default",
      "setting_name": "",
      "score": 89.6,
      "unit": "%",
      "printed": "89.6",
      "reported_by": "moonshot",
      "run_by": null,
      "label": "Kimi K2.6",
      "source": "https://www.kimi.ai/blog/kimi-k2-6",
      "source_title": "Kimi K2.6: Advancing Open-Source Coding",
      "source_date": "2026-04-20",
      "where": "Benchmark Table (HTML), row 'LiveCodeBench (v6)', column 'Kimi K2.6'",
      "own": true,
      "primary": true,
      "parallel": false,
      "notes": "",
      "caveat": "",
      "human_verified": false,
      "human_verified_on": null,
      "check": "corroborated",
      "corroborated": true,
      "corroborated_on": "2026-10-28",
      "corroborated_by": [
        {
          "url": "https://huggingface.co/deepseek-ai/DeepSeek-V4-Pro",
          "name": "DeepSeek V4 Preview card",
          "title": "DeepSeek-V4-Pro model card",
          "lab": "deepseek",
          "date": "2026-04-24",
          "printed": "89.6",
          "where": "DeepSeek card, frontier comparison table, column \"K2.6 Thinking\", row \"LiveCodeBench (Pass@1)\""
        }
      ],
      "source_capture": "https://frontierlog.ai/captures/a8b4967f3aab7948.png"
    },
    {
      "id": "qwen3-7-plus:livecodebench-v6:default:1",
      "model": "qwen3-7-plus",
      "model_name": "Qwen3.7-Plus",
      "model_date": "2026-06-01",
      "event": "https://frontierlog.ai/event/2026-06-01-qwen3-7-plus",
      "benchmark": "livecodebench-v6",
      "benchmark_name": "LiveCodeBench v6",
      "setting": "default",
      "setting_name": "",
      "score": 89.6,
      "unit": "%",
      "printed": "89.6",
      "reported_by": "alibaba",
      "run_by": null,
      "label": "Qwen3.7-Plus",
      "source": "https://qwen.ai/blog?id=qwen3.7-plus",
      "source_title": "Qwen3.7-Plus: Multimodal Agent Intelligence",
      "source_date": "2026-06-01",
      "where": "'Performance' section, language table (HTML), STEM & Reasoning block, row 'LiveCodeBench', column 'Qwen3.7-Plus'",
      "own": true,
      "primary": true,
      "parallel": false,
      "notes": "",
      "caveat": "",
      "human_verified": true,
      "human_verified_on": "2026-10-28",
      "check": "human",
      "corroborated": true,
      "corroborated_on": "2026-10-28",
      "corroborated_by": [
        {
          "url": "https://huggingface.co/Qwen/Qwen3.8-Flash-Next",
          "name": "Qwen3.8-Flash-Next card",
          "title": "Qwen/Qwen3.8-Flash-Next model card",
          "lab": "alibaba",
          "date": "2026-08-26",
          "printed": "89.6",
          "where": "Qwen3.8-Flash-Next card, language table, row Competitive coding / LiveCodeBench v6, column Qwen3.7-Plus"
        }
      ],
      "source_capture": "https://frontierlog.ai/captures/de169a5a059c61f9.png"
    },
    {
      "id": "gemma-4-12b:livecodebench-v6:default:1",
      "model": "gemma-4-12b",
      "model_name": "Gemma 4 12B",
      "model_date": "2026-06-03",
      "event": "https://frontierlog.ai/event/2026-06-03-gemma-4-12b",
      "benchmark": "livecodebench-v6",
      "benchmark_name": "LiveCodeBench v6",
      "setting": "default",
      "setting_name": "",
      "score": 72.0,
      "unit": "%",
      "printed": "72",
      "reported_by": "google",
      "run_by": null,
      "label": "Gemma 4 12B",
      "source": "https://blog.google/innovation-and-ai/technology/developers-tools/introducing-gemma-4-12b/",
      "source_title": "Introducing Gemma 4 12B: a unified, encoder-free multimodal model",
      "source_date": "2026-06-03",
      "where": "'Gemma 4 12B Benchmarks' bar chart (image), group 'LiveCode Bench', value label over the 'Gemma 4 12B' bar (middle bar)",
      "own": true,
      "primary": true,
      "parallel": false,
      "notes": "",
      "caveat": "",
      "human_verified": false,
      "human_verified_on": null,
      "check": "corroborated",
      "corroborated": true,
      "corroborated_on": "2026-10-28",
      "corroborated_by": [
        {
          "url": "https://huggingface.co/google/gemma-4-12B-it",
          "name": "Gemma 4 12B card",
          "title": "google/gemma-4-12B-it model card",
          "lab": "google",
          "date": "2026-06-03",
          "printed": "72.0%",
          "where": "'Benchmark Results' table (instruction-tuned), row 'LiveCodeBench v6', column 'Gemma 4 12B Unified'"
        },
        {
          "url": "https://ai.google.dev/gemma/docs/core/model_card_4",
          "name": "Gemma 4 card",
          "title": "Gemma 4 model card",
          "lab": "google",
          "date": "2026-04-02",
          "printed": "72.0%",
          "where": "Benchmark Results table (instruction-tuned), row 'LiveCodeBench v6', column 'Gemma 4 12B Unified'"
        }
      ],
      "source_image": "https://frontierlog.ai/sources/e419a76ecd79078d.webp",
      "source_image_original": "https://storage.googleapis.com/gweb-uniblog-publish-prod/images/1920x1080_xMVEyWv.width-1200.format-webp.webp"
    },
    {
      "id": "nemotron-3-ultra-550b-a55b:livecodebench-v6:default:1",
      "model": "nemotron-3-ultra-550b-a55b",
      "model_name": "Nemotron 3 Ultra 550B-A55B",
      "model_date": "2026-06-04",
      "event": "https://frontierlog.ai/event/2026-06-04-nemotron-3-ultra",
      "benchmark": "livecodebench-v6",
      "benchmark_name": "LiveCodeBench v6",
      "setting": "default",
      "setting_name": "",
      "score": 89.0,
      "unit": "%",
      "printed": "89.0",
      "reported_by": "nvidia",
      "run_by": null,
      "label": "N-3-Ultra",
      "source": "https://research.nvidia.com/labs/nemotron/files/NVIDIA-Nemotron-3-Ultra-Technical-Report.pdf",
      "source_title": "Nemotron 3 Ultra: Open, Efficient Mixture-of-Experts Hybrid Mamba-Transformer Model for Agentic Reasoning",
      "source_date": "2026-06-04",
      "where": "Table 10 (PDF page 38), row 'LiveCodeBench (v6)', column 'N-3-Ultra'",
      "own": true,
      "primary": true,
      "parallel": false,
      "notes": "",
      "caveat": "",
      "human_verified": false,
      "human_verified_on": null,
      "check": "corroborated",
      "corroborated": true,
      "corroborated_on": "2026-10-19",
      "corroborated_by": [
        {
          "url": "https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16/blob/77df655d5e9f8362164ed14dd8b48f8bce657498/README.md",
          "name": "Nemotron 3 Ultra card",
          "title": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16 model card (revision of 2026-08-24)",
          "lab": "nvidia",
          "date": "2026-06-04",
          "printed": "89.0",
          "where": "Benchmarks table, 'N-3-Ultra 550B-A55B' column, LiveCodeBench (v6) row"
        }
      ],
      "source_capture": "https://frontierlog.ai/captures/a737368a281c4576.png"
    },
    {
      "id": "nemotron-3-ultra-550b-a55b:livecodebench-v6:default:2",
      "model": "nemotron-3-ultra-550b-a55b",
      "model_name": "Nemotron 3 Ultra 550B-A55B",
      "model_date": "2026-06-04",
      "event": "https://frontierlog.ai/event/2026-06-04-nemotron-3-ultra",
      "benchmark": "livecodebench-v6",
      "benchmark_name": "LiveCodeBench v6",
      "setting": "default",
      "setting_name": "",
      "score": 89.21,
      "unit": "%",
      "printed": "89.21",
      "reported_by": "nvidia",
      "run_by": null,
      "label": "N3-Ultra 550B-A55B",
      "source": "https://research.nvidia.com/labs/nemotron/files/NVIDIA-Nemotron-3.5-Super-and-Lightning-Technical-Report.pdf",
      "source_title": "Nemotron 3.5 Super and Lightning: Open, Efficient Mixture-of-Experts Hybrid Mamba-Transformer Models for Agentic AI",
      "source_date": "2026-10-15",
      "where": "Table 10 (PDF page 23), row 'LiveCodeBench v6', column 'N3-Ultra 550B-A55B' (second column)",
      "own": false,
      "primary": false,
      "parallel": false,
      "notes": "problems from Aug 2024 to May 2025; average of 3 samples",
      "caveat": "",
      "human_verified": false,
      "human_verified_on": null,
      "check": "corroborated",
      "corroborated": true,
      "corroborated_on": "2026-10-19",
      "corroborated_by": [
        {
          "url": "https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16/blob/1f8953f7f0935358e3605150d22bc15df062061c/README.md",
          "name": "Nemotron 3.5 Super VL and Bolt card",
          "title": "nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16 model card (revision of 2026-10-16)",
          "lab": "nvidia",
          "date": "2026-10-15",
          "printed": "89.2",
          "where": "evaluation table, 'Nemotron 3 Ultra' column, LiveCodeBench v6 row"
        }
      ],
      "source_capture": "https://frontierlog.ai/captures/ee443a24cf8af62e.png"
    },
    {
      "id": "diffusiongemma-26b-a4b:livecodebench-v6:default:1",
      "model": "diffusiongemma-26b-a4b",
      "model_name": "DiffusionGemma 26B A4B",
      "model_date": "2026-06-10",
      "event": "https://frontierlog.ai/event/2026-06-10-diffusiongemma",
      "benchmark": "livecodebench-v6",
      "benchmark_name": "LiveCodeBench v6",
      "setting": "default",
      "setting_name": "",
      "score": 69.1,
      "unit": "%",
      "printed": "69.1%",
      "reported_by": "google",
      "run_by": null,
      "label": "DiffusionGemma 26B A4B",
      "source": "https://blog.google/innovation-and-ai/technology/developers-tools/diffusion-gemma-faster-text-generation/",
      "source_title": "DiffusionGemma: 4x faster text generation",
      "source_date": "2026-06-10",
      "where": "'DiffusionGemma' benchmark bar chart (image), panel 'LiveCodeBench v6', value label over the 'Diffusion Gemma 26B A4B' bar (left bar)",
      "own": true,
      "primary": true,
      "parallel": false,
      "notes": "",
      "caveat": "",
      "human_verified": false,
      "human_verified_on": null,
      "check": "corroborated",
      "corroborated": true,
      "corroborated_on": "2026-10-28",
      "corroborated_by": [
        {
          "url": "https://huggingface.co/google/diffusiongemma-26B-A4B-it",
          "name": "DiffusionGemma card",
          "title": "google/diffusiongemma-26B-A4B-it model card",
          "lab": "google",
          "date": "2026-06-10",
          "printed": "69.1%",
          "where": "'Benchmark Results' table, row 'LiveCodeBench v6', column 'DiffusionGemma 26B A4B' (instruction-tuned, Entropy Bound sampler)"
        }
      ],
      "source_image": "https://frontierlog.ai/sources/83cdd78e30efb08d.webp",
      "source_image_original": "https://storage.googleapis.com/gweb-uniblog-publish-prod/images/diffusiongemma__benchmark__bar_l.width-1200.format-webp.webp"
    },
    {
      "id": "qwen3-8-flash-next:livecodebench-v6:default:1",
      "model": "qwen3-8-flash-next",
      "model_name": "Qwen3.8-Flash-Next",
      "model_date": "2026-08-26",
      "event": "https://frontierlog.ai/event/2026-08-26-qwen3-8-flash-next",
      "benchmark": "livecodebench-v6",
      "benchmark_name": "LiveCodeBench v6",
      "setting": "default",
      "setting_name": "",
      "score": 91.9,
      "unit": "%",
      "printed": "91.9",
      "reported_by": "alibaba",
      "run_by": null,
      "label": "Qwen3.8-Flash-Next",
      "source": "https://qwen.ai/blog?id=qwen3.8-flash-next",
      "source_title": "Qwen3.8-Flash-Next: A New Architecture, Towards Ultimate Cost-Efficiency",
      "source_date": "2026-08-26",
      "where": "'Language' table (HTML) under 'Performance', General block, row 'LiveCodeBench v6' (Competitive coding), column 'Qwen3.8-Flash-Next'",
      "own": true,
      "primary": true,
      "parallel": false,
      "notes": "",
      "caveat": "",
      "human_verified": true,
      "human_verified_on": "2026-10-28",
      "check": "human",
      "corroborated": true,
      "corroborated_on": "2026-10-28",
      "corroborated_by": [
        {
          "url": "https://huggingface.co/Qwen/Qwen3.8-Flash-Next",
          "name": "Qwen3.8-Flash-Next card",
          "title": "Qwen/Qwen3.8-Flash-Next model card",
          "lab": "alibaba",
          "date": "2026-08-26",
          "printed": "91.9",
          "where": "language table, row Competitive coding / LiveCodeBench v6, column Qwen3.8-Flash-Next"
        }
      ],
      "source_capture": "https://frontierlog.ai/captures/1ce175beb0b20cf1.png"
    },
    {
      "id": "qwen3-8-omni-flash:livecodebench-v6:default:1",
      "model": "qwen3-8-omni-flash",
      "model_name": "Qwen3.8-Omni-Flash",
      "model_date": "2026-09-18",
      "event": "https://frontierlog.ai/event/2026-09-18-qwen3-8-omni-flash",
      "benchmark": "livecodebench-v6",
      "benchmark_name": "LiveCodeBench v6",
      "setting": "default",
      "setting_name": "",
      "score": 92.6,
      "unit": "%",
      "printed": "92.6",
      "reported_by": "alibaba",
      "run_by": null,
      "label": "Qwen3.8-Omni-Flash",
      "source": "https://qwen.ai/blog?id=qwen3.8-omni-flash",
      "source_title": "Qwen3.8-Omni-Flash: Omni Senses. Agentic Delivery.",
      "source_date": "2026-09-18",
      "where": "'Text' table (HTML), General Text Capabilities block, row 'LiveCodeBench v6' (Competitive coding), column 'Qwen3.8-Omni-Flash'",
      "own": true,
      "primary": true,
      "parallel": false,
      "notes": "",
      "caveat": "",
      "human_verified": true,
      "human_verified_on": "2026-10-28",
      "check": "human",
      "corroborated": false,
      "corroborated_on": null,
      "corroborated_by": [],
      "source_capture": "https://frontierlog.ai/captures/0f628ae53818eafc.png"
    },
    {
      "id": "nemotron-3-5-super-vl-120b-a12b:livecodebench-v6:default:1",
      "model": "nemotron-3-5-super-vl-120b-a12b",
      "model_name": "Nemotron 3.5 Super VL 120B-A12B",
      "model_date": "2026-10-15",
      "event": "https://frontierlog.ai/event/2026-10-15-nemotron-3-5-super-vl-bolt",
      "benchmark": "livecodebench-v6",
      "benchmark_name": "LiveCodeBench v6",
      "setting": "default",
      "setting_name": "",
      "score": 88.22,
      "unit": "%",
      "printed": "88.22",
      "reported_by": "nvidia",
      "run_by": null,
      "label": "Nemotron 3.5 SuperGA BF16",
      "source": "https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16/blob/8350289dda26a68b7d7f5c8b8ee7f815b5d1bca9/README.md",
      "source_title": "nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16 model card (revision of 2026-10-15)",
      "source_date": "2026-10-15",
      "where": "model card (launch-day revision), 'Benchmarks' table, row 'LiveCodeBench v6', column 'Nemotron 3.5 SuperGA BF16'",
      "own": true,
      "primary": true,
      "parallel": false,
      "notes": "",
      "caveat": "",
      "human_verified": false,
      "human_verified_on": null,
      "check": "corroborated",
      "corroborated": true,
      "corroborated_on": "2026-10-28",
      "corroborated_by": [
        {
          "url": "https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-NVFP4/blob/8a115b9d966adce8c059a7bcf47f49a1e57957d8/README.md",
          "name": "Nemotron 3.5 Super VL and Bolt card",
          "title": "nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-NVFP4 model card (launch-day revision)",
          "lab": "nvidia",
          "date": "2026-10-15",
          "printed": "88.22",
          "where": "NVFP4 model card (launch-day revision), benchmark table, LiveCodeBench v6 row, 'Nemotron 3.5 Super VL BF16' column"
        },
        {
          "url": "https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16/blob/1f8953f7f0935358e3605150d22bc15df062061c/README.md",
          "name": "Nemotron 3.5 Super VL and Bolt card",
          "title": "nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16 model card (revision of 2026-10-16)",
          "lab": "nvidia",
          "date": "2026-10-15",
          "printed": "88.22",
          "where": "Later revision (2026-10-16) of the same BF16 card as `not`: benchmark table, LiveCodeBench v6 row, 'Nemotron 3.5 SuperGA BF16' column"
        },
        {
          "url": "https://research.nvidia.com/labs/nemotron/files/NVIDIA-Nemotron-3.5-Super-and-Lightning-Technical-Report.pdf",
          "name": "Nemotron 3.5 Super VL and Bolt card",
          "title": "Nemotron 3.5 Super and Lightning: Open, Efficient Mixture-of-Experts Hybrid Mamba-Transformer Models for Agentic AI",
          "lab": "nvidia",
          "date": "2026-10-15",
          "printed": "88.22",
          "where": "Table 10 (Evaluation suite for Nemotron 3.5 Super), LiveCodeBench v6 row, N3.5-Super 120B-A12B column; same value in Table 23 (LiveCodeBench v6, BF16 acc.)"
        }
      ],
      "source_capture": "https://frontierlog.ai/captures/99b6b838a462495b.png"
    }
  ]
}