EvalPlus: leaderboard
EvalPlus leaderboard aggregating HumanEval+ and MBPP+ code-generation pass@1 scores.
Metric: EvalPlus Avg. (imported). Source: evalplus.github.io. Source: benchmarklist.com. Saturation forecast: Estimated already saturated. 25 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | O1 Preview (2024-09-12) | 84.6 | #197 |
| 2 | O1 Mini (2024-09-12) | 83.9 | #403 |
| 3 | Qwen 2.5 Coder 32B Instruct | 82.1 | #528 |
| 4 | GPT-4o (2024-08-06) | 79.7 | #336 |
| 5 | Claude 3.5 Sonnet (20240620) | 78 | #299 |
| 6 | GPT-4o Mini (2024-07-18) | 77.85 | #558 |
| 7 | GPT-4 Turbo (Preview) | 77.5 | #397 |
| 8 | Gemini 1.5 Pro (002) | 76.95 | #349 |
| 9 | Claude 3 Opus | 75.35 | #571 |
| 10 | OpenCoder-8B-Instruct | 74.4 | #1338 |
| 11 | CodeQwen1.5-7B Chat | 73.85 | #1144 |
| 12 | Grok Beta | 73.05 | #487 |
| 13 | Gemini 1.5 Flash (002) | 71.55 | #583 |
| 14 | Llama 3 70B Instruct | 70.5 | #634 |
| 15 | GPT-3.5 Turbo (1106) | 70.2 | #961 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=evalplus · How It Works · Data refreshed daily, snapshot 2026-10-20.