Vibeleaderboard
← Benchmark guide

Model results

WeirdML

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.

Official source
← All evaluation groups

WeirdML

Source
WeirdML via Epoch AI
Version
v2
Measurement date
Not reported
Metric
% · Higher is better
Evaluation scope
result date not supplied

Positions are withheld because at least one result lacks explicit independent provenance or a documented evaluation scope. Missing metadata is shown as unreported; matching imported metadata is not independent verification of the protocol.

  1. Claude Opus 5

    anthropic/claude-opus-5

    86.3%

    View modelEvidence aggregated

  2. GPT-5.3-Codex

    openai/gpt-5.3-codex-20260224

    79.3%

    View modelEvidence aggregated

  3. Claude Opus 4.7

    anthropic/claude-opus-4-7

    76.4%

    View modelEvidence aggregated

  4. Gemini 3.1 Pro Preview

    google/gemini-3.1-pro-preview-20260219

    72.1%

    View modelEvidence aggregated

  5. Claude Opus 4.6

    anthropic/claude-opus-4-6

    65.9%

    View modelEvidence aggregated

  6. Gemini 3 Flash Preview

    google/gemini-3-flash-preview-20251217

    61.6%

    View modelEvidence aggregated

  7. GLM 5.1

    z-ai/glm-5.1-20260406

    57.1%

    View modelEvidence aggregated

  8. Kimi K2.6

    moonshotai/kimi-k2.6-20260420

    55.9%

    View modelEvidence aggregated

  9. GPT-5 Codex

    openai/gpt-5-codex

    54.5%

    View modelEvidence aggregated

  10. Kimi K2.7 Code

    moonshotai/kimi-k2.7-code-20260612

    54.1%

    View modelEvidence aggregated

  11. Gemma 4 31B

    google/gemma-4-31b-it-20260402

    52.3%

    View modelEvidence aggregated

  12. Grok 4.20

    x-ai/grok-4.20-20260309

    52.3%

    View modelEvidence aggregated

  13. Gemini 3.1 Flash Lite

    google/gemini-3.1-flash-lite-20260507

    52.2%

    View modelEvidence aggregated

  14. GLM 5

    z-ai/glm-5-20260211

    48.2%

    View modelEvidence aggregated

  15. Claude Sonnet 4.5

    anthropic/claude-sonnet-4-5

    46.7%

    View modelEvidence aggregated

  16. Kimi K2.5

    moonshotai/kimi-k2.5-0127

    45.6%

    View modelEvidence aggregated

  17. Claude Haiku 4.5

    anthropic/claude-haiku-4-5

    45.4%

    View modelEvidence aggregated

  18. Claude Sonnet 4

    anthropic/claude-sonnet-4

    43.9%

    View modelEvidence aggregated

  19. Nemotron 3 Ultra

    nvidia/nemotron-3-ultra-550b-a55b-20260604

    43.5%

    View modelEvidence aggregated

  20. Mercury 2

    inception/mercury-2-20260304

    43.2%

    View modelEvidence aggregated

  21. Kimi K2 Thinking

    moonshotai/kimi-k2-thinking-20251106

    42.8%

    View modelEvidence aggregated

  22. R1 0528

    deepseek/deepseek-r1-0528

    41.6%

    View modelEvidence aggregated

  23. Qwen3 Coder 480B A35B

    qwen/qwen3-coder-480b-a35b-07-25

    41.2%

    View modelEvidence aggregated

  24. Qwen3.5-27B

    qwen/qwen3.5-27b-20260224

    39.5%

    View modelEvidence aggregated

  25. Kimi K2 0711

    moonshotai/kimi-k2

    39.4%

    View modelEvidence aggregated

  26. GPT-4.1

    openai/gpt-4.1-2025-04-14

    39%

    View modelEvidence aggregated

  27. Qwen3 235B A22B Instruct 2507

    qwen/qwen3-235b-a22b-07-25

    38.7%

    View modelEvidence aggregated

  28. Nemotron 3 Super

    nvidia/nemotron-3-super-120b-a12b-20230311

    38%

    View modelEvidence aggregated

  29. GPT-4.1 Mini

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

    37.6%

    View modelEvidence aggregated

  30. DeepSeek V3.1

    deepseek/deepseek-chat-v3.1

    37.5%

    View modelEvidence aggregated

  31. Qwen3 235B A22B

    qwen/qwen3-235b-a22b-04-28

    37.3%

    View modelEvidence aggregated

  32. MiniMax M2.7

    minimax/minimax-m2.7-20260318

    37%

    View modelEvidence aggregated

  33. Kimi K2 0905

    moonshotai/kimi-k2-0905

    36.7%

    View modelEvidence aggregated

  34. R1

    deepseek/deepseek-r1

    36.5%

    View modelEvidence aggregated

  35. DeepSeek V3 0324

    deepseek/deepseek-chat-v3-0324

    36.1%

    View modelEvidence aggregated

  36. Gemma 4 26B A4B

    google/gemma-4-26b-a4b-it-20260403

    35.2%

    View modelEvidence aggregated

  37. Qwen3.6 35B A3B

    qwen/qwen3.6-35b-a3b-20260415

    34.5%

    View modelEvidence aggregated

  38. Qwen3 Coder Next

    qwen/qwen3-coder-next-2025-02-03

    34.4%

    View modelEvidence aggregated

  39. Qwen3 30B A3B

    qwen/qwen3-30b-a3b-04-28

    29.8%

    View modelEvidence aggregated

  40. GPT-4o (2024-11-20)

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

    25.1%

    View modelEvidence aggregated

  41. Llama 4 Maverick

    meta/llama-4-maverick-17b-128e-instruct

    24.5%

    View modelEvidence aggregated

  42. GPT-4.1 Nano

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

    19%

    View modelEvidence aggregated

  43. Qwen2.5 72B Instruct

    qwen/qwen-2.5-72b-instruct

    16%

    View modelEvidence aggregated

  44. Llama 3.3 70B Instruct

    meta/llama-3.3-70b-instruct

    14.4%

    View modelEvidence aggregated

  45. GPT-4o-mini (2024-07-18)

    openai/gpt-4o-mini-2024-07-18

    11.8%

    View modelEvidence aggregated

  46. Claude 3 Haiku

    anthropic/claude-3-haiku

    9.8%

    View modelEvidence aggregated

  47. Llama 3.1 70B Instruct

    meta/llama-3.1-70b-instruct

    9%

    View modelEvidence aggregated

  48. Mixtral 8x22B Instruct

    mistralai/mixtral-8x22b-instruct

    3.2%

    View modelEvidence aggregated

  49. Llama 3.1 8B Instruct

    meta/llama-3.1-8b-instruct

    1.7%

    View modelEvidence aggregated
WeirdML model results | VibeLeaderboard