Vibeleaderboard
← Benchmark guide

Model results

FrontierMath

Choose a shared evaluation protocol to compare model scores. Versions, sources and configurations stay separate.

Atlas snapshot captured 10/19/2026, 10:20:28 AM UTC. This is an import timestamp, not the evaluation date. Coverage may be incomplete.

Reviewed recommendation and limits

Use the relevant tier and private/public split for difficult mathematical reasoning.

Each tier, release and public/private split separately. Imported groups below may use different protocols; this review does not certify each run. Difficulty is specialized; tier, dataset release, tools and grader must match before comparing scores.

Publisher results Epoch benchmarking methods
← All evaluation groups

FrontierMath 2025-02-28 · private

Source
Epoch AI Benchmarking Hub
Version
2025-02-28-private
Measurement date
Not reported
Metric
% · Higher is better
Evaluation scope
Epoch AI run · task v1 · superseded by FrontierMath Tiers 1–3 v2 · private · result date not supplied

Exact score ties share a position. Positions do not establish statistical significance. Missing metadata is shown as unreported; matching imported metadata is not independent verification of the protocol.

  1. #1Kimi K2.6

    moonshotai/kimi-k2.6-20260420

    39%

    View modelEvidence independent

  2. #2Claude Opus 4.6

    anthropic/claude-opus-4-6

    38.3%

    View modelEvidence independent

  3. #3Gemini 3.1 Pro Preview

    google/gemini-3.1-pro-preview-20260219

    36.9%

    View modelEvidence independent

  4. #4Gemini 3 Flash Preview

    google/gemini-3-flash-preview-20251217

    35.6%

    View modelEvidence independent

  5. #5GLM 5.1

    z-ai/glm-5.1-20260406

    33.4%

    View modelEvidence independent

  6. #6Qwen3.6 Plus

    qwen/qwen3.6-plus-04-02

    26.2%

    View modelEvidence independent

  7. #7Qwen3.6 Max Preview

    qwen/qwen3.6-max-preview-20260420

    23.1%

    View modelEvidence independent

  8. #8Kimi K2 Thinking

    moonshotai/kimi-k2-thinking-20251106

    21.4%

    View modelEvidence independent

  9. #9Claude Opus 4.5

    anthropic/claude-opus-4-5

    20.7%

    View modelEvidence independent

  10. #10GLM 5

    z-ai/glm-5-20260211

    16.4%

    View modelEvidence independent

  11. #11Gemini 2.5 Pro

    google/gemini-2.5-pro

    14.1%

    View modelEvidence independent

  12. #12Gemini 2.5 Pro Preview 06-05

    google/gemini-2.5-pro-preview-06-05

    10.3%

    View modelEvidence independent

  13. #12Qwen3.6 Flash

    qwen/qwen3.6-flash

    10.3%

    View modelEvidence independent

  14. #14Claude Sonnet 4.5

    anthropic/claude-sonnet-4-5

    9.3%

    View modelEvidence independent

  15. #15Qwen3 235B A22B Thinking 2507

    qwen/qwen3-235b-a22b-thinking-2507

    8.5%

    View modelEvidence independent

  16. #16Qwen3.5-Flash

    qwen/qwen3.5-flash-20260224

    6.2%

    View modelEvidence independent

  17. #17Claude Opus 4.1

    anthropic/claude-opus-4-1

    5.9%

    View modelEvidence independent

  18. #18GPT-4.1

    openai/gpt-4.1-2025-04-14

    5.5%

    View modelEvidence independent

  19. #19Gemini 2.5 Flash

    google/gemini-2.5-flash

    4.8%

    View modelEvidence independent

  20. #20Claude Opus 4

    anthropic/claude-opus-4

    4.5%

    View modelEvidence independent

  21. #20GPT-4.1 Mini

    openai/gpt-4.1-mini-2025-04-14

    4.5%

    View modelEvidence independent

  22. #22Claude Haiku 4.5

    anthropic/claude-haiku-4-5

    4.1%

    View modelEvidence independent

  23. #22Claude Sonnet 4

    anthropic/claude-sonnet-4

    4.1%

    View modelEvidence independent

  24. #24GLM 4.6

    z-ai/glm-4.6

    3.8%

    View modelEvidence independent

  25. #25GLM 4.7

    z-ai/glm-4.7-20251222

    2.4%

    View modelEvidence independent

  26. #26DeepSeek V3

    deepseek/deepseek-chat-v3

    1.7%

    View modelEvidence independent

  27. #26Qwen-Plus

    qwen/qwen-plus-2025-01-25

    1.7%

    View modelEvidence independent

  28. #28GPT-4.1 Nano

    openai/gpt-4.1-nano-2025-04-14

    1%

    View modelEvidence independent

  29. #29GPT-4o (2024-08-06)

    openai/gpt-4o-2024-08-06

    0.3%

    View modelEvidence independent

  30. #29GPT-4o (2024-11-20)

    openai/gpt-4o-2024-11-20

    0.3%

    View modelEvidence independent

  31. #31Llama 4 Scout

    meta/llama-4-scout-17b-16e-instruct

    0%

    View modelEvidence independent
FrontierMath model results | VibeLeaderboard