Benchmark Leaderboard
Top models by HumanEval score
Sorts are only valid inside this benchmark. A score gap of this benchmark does not map to the same gap on another benchmark, so validate any short-list by checking peer rows and model context/price fit.
| # | Model | Score | Version | Source |
|---|---|---|---|---|
| 1 | Claude Sonnet 4.6 | 98 | HumanEval (pass@1) | https://automatio.ai/models/claude-sonnet-4-6 |
| 2 | o3 | 96.7 | 2025-04 | https://openai.com/index/introducing-o3-and-o4-mini/ |
| 3 | Claude Opus 4.6 | 95 | HumanEval (pass@1) | https://automatio.ai/models/claude-opus-4-6 |
| 4 | Grok-3 | 94.5 | — | https://x.ai/blog/grok-3 |
| 5 | GPT-5.5 | 94.2 | — | https://openai.com/index/introducing-gpt-5-5/ |
| 6 | Gemini 3.1 Pro Preview | 94 | HumanEval (pass@1) | https://automatio.ai/models/gemini-3-1-pro |
| 7 | Gemini 2.5 Pro | 93.1 | 2025-03 | https://storage.googleapis.com/deepmind-media/gemini/gemini_v2_5_report.pdf |
| 8 | Claude 3.7 Sonnet | 93 | 2025-02 | https://www.anthropic.com/news/claude-3-7-sonnet |
| 9 | GPT-4.1 | 92.9 | 2025-04 | https://openai.com/index/gpt-4-1/ |
| 10 | Qwen2.5-Coder-32B-Instruct | 92.7 | 2024-11 | https://qwenlm.github.io/blog/qwen2.5-coder-family/ |
| 11 | Qwen3-235B-A22B | 92.7 | 2025-04 | https://qwenlm.github.io/blog/qwen3/ |
| 12 | Claude 3.5 Sonnet | 92 | pass@1 | https://crfm.stanford.edu/helm/classic/latest/ |
| 13 | Kimi K2.6 | 92 | — | https://moonshotai.github.io/Kimi-K2/ |
| 14 | Mistral Large 3 675B Instruct | 92 | — | https://mistral.ai/news/mistral-large-3/ |
| 15 | Gemini 3.5 Flash | 92 | — | https://o-mega.ai/articles/gemini-3-5-flash-benchmarks-cost-and-guide |
| 16 | OLMo 3 32B Think | 91.4 | HumanEval (pass@1) | https://huggingface.co/allenai/Olmo-3-32B-Think |
| 17 | GPT-4o (05-13) | 90.2 | pass@1 | https://crfm.stanford.edu/helm/classic/latest/ |
| 18 | Gemini 2.5 Flash | 90.1 | 2025-05 | https://storage.googleapis.com/deepmind-media/gemini/gemini_v2_5_report.pdf |
| 19 | DeepSeek R1 | 89.9 | 2025-01 | https://arxiv.org/abs/2501.12948 |
| 20 | Granite 4.1 30B | 89.63 | pass@1, instruct model (pass@1) | https://huggingface.co/blog/ibm-granite/granite-4-1 |
Interpretation
Trust when:
- Use top rows for shortlisting, then validate provider coverage and feature fit before selecting a model.
- Cross-benchmark consistency is a stronger signal than one-off score dominance.
Don't trust when:
- Don't compare raw score gaps across different benchmark scales as equivalent.
- Missing rows in the heatmap can hide benchmark blind spots.
- No trend data means this is a snapshot; expect drift in fresh refresh cycles.
Cross-benchmark heatmap
Compare top models across selected benchmark families to spot strengths and weak spots. On narrow screens, scroll horizontally to inspect every benchmark.
| Model | GPQA | HellaSwag | HumanEval | MMLU |
|---|---|---|---|---|
| Claude Sonnet 4.6 | 89.9 | — | 98.0 | 89.3 |
| o3 | 87.7 | — | 96.7 | — |
| Claude Opus 4.6 | 91.3 | — | 95.0 | 91.1 |
| Grok-3 | 84.6 | — | 94.5 | — |
| GPT-5.5 | 93.6 | — | 94.2 | 92.4 |
| Gemini 3.1 Pro Preview | 94.3 | — | 94.0 | 98.0 |