EvalPlus (HumanEval+ & MBPP+) — leaderboard
Metric: Pass@1 avg (%). Source: evalplus.github.io. 125 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-4 Turbo | 86.6 |
| 2 | O1 Mini (2024-09-12) | 83.9 |
| 3 | Qwen 2.5 Coder 32B Instruct | 82.1 |
| 4 | GPT-4o (2024-08-06) | 79.7 |
| 5 | GPT-4 | 79.3 |
| 6 | Claude 3.5 Sonnet (20240620) | 78 |
| 7 | GPT-4o Mini (2024-07-18) | 77.8 |
| 8 | GPT-4 Turbo (Preview) | 77.5 |
| 9 | Gemini 1.5 Pro (002) | 76.9 |
| 10 | Claude 3 Opus | 75.3 |
| 11 | OpenCoder-8B-Instruct | 74.4 |
| 12 | Grok Beta | 73 |
| 13 | Gemini 1.5 Flash (002) | 71.5 |
| 14 | Llama 3 70B Instruct | 70.5 |
| 15 | GPT-3.5 Turbo (1106) | 70.2 |
Interactive version: theaggregate.ai/benchmark?slug=evalplus-humaneval-plus-mbpp-plus · How the rankings work · Data refreshed daily, snapshot 2026-07-22.