MBPP+: leaderboard
378 hand-verified Python problems drawn from MBPP-sanitized, each with 35x more test cases than the original MBPP, built by the EvalPlus project (NeurIPS 2023); pass@1 with greedy decoding.
Metric: MBPP+ pass@1 (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) | 80.2 | #197 |
| 2 | O1 Mini (2024-09-12) | 78.8 | #403 |
| 3 | Qwen 2.5 Coder 32B Instruct | 77 | #528 |
| 4 | Gemini 1.5 Pro (002) | 74.6 | #349 |
| 5 | Claude 3.5 Sonnet (20240620) | 74.3 | #299 |
| 6 | Claude 3 Opus | 73.3 | #571 |
| 7 | GPT-4 Turbo (Preview) | 73.3 | #397 |
| 8 | GPT-4o Mini (2024-07-18) | 72.2 | #558 |
| 9 | GPT-4o (2024-08-06) | 72.2 | #336 |
| 10 | OpenCoder-8B-Instruct | 71.4 | #1338 |
| 11 | GPT-3.5 Turbo (1106) | 69.7 | #961 |
| 12 | Claude 3 Sonnet | 69.3 | #775 |
| 13 | Llama 3 70B Instruct | 69 | #634 |
| 14 | Magicoder-S-DS-6.7B | 69 | #1256 |
| 15 | CodeQwen1.5-7B Chat | 69 | #1144 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=mbpp-plus · How It Works · Data refreshed daily, snapshot 2026-10-20.