EvoEval Combine — leaderboard
Metric: Pass@1 (%). Source: evo-eval.github.io. 51 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-4 | 53 |
| 2 | GPT-4 Turbo | 45 |
| 3 | Claude 3 Haiku | 25 |
| 4 | Gemini 1.0 Pro | 23 |
| 5 | Claude 2 | 19 |
| 6 | deepseek-coder-6.7B-instruct | 18 |
| 7 | Phi-2 | 14 |
| 8 | Qwen-14B | 13 |
| 9 | Mixtral 8x7B Instruct | 9 |
| 10 | deepseek-coder-1.3B-instruct | 9 |
| 11 | starcoder2-15B | 5 |
| 12 | Qwen-7B | 4 |
| 13 | Mistral 7B Instruct | 3 |
| 14 | Mistral 7B | 3 |
| 15 | gemma-7B | 2 |
Interactive version: theaggregate.ai/benchmark?slug=evoeval-combine · How the rankings work · Data refreshed daily, snapshot 2026-07-22.