Vector Eval - HumanEval: leaderboard
Metric: Mean Score (%). Source: huggingface.co. 12 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | O3 Mini | 98.17 |
| 2 | O1 | 96.95 |
| 3 | DeepSeek R1 | 95.68 |
| 4 | Claude 3.5 Sonnet | 94.51 |
| 5 | Qwen3 Coder Next | 91.46 |
| 6 | GPT-4o | 90.85 |
| 7 | Gemini 1.5 Pro | 87.2 |
| 8 | Mistral Large 2 (Jul) | 86.59 |
| 9 | GPT-4o Mini | 85.98 |
| 10 | Llama 3.1 70B Instruct | 78.66 |
| 11 | Gemini 1.5 Flash | 74.39 |
| 12 | Command-R+ | 62.2 |
Interactive version: theaggregate.ai/benchmark?slug=vector-eval-humaneval · How It Works · Data refreshed daily, snapshot 2026-09-05.