EvoEval Verbose — leaderboard
Metric: Pass@1 (%). Source: evo-eval.github.io. 17 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-4 Turbo | 79.27 |
| 2 | deepseek-coder-6.7B-instruct | 72.56 |
| 3 | Claude 3 Haiku | 71.34 |
| 4 | GPT-4 | 70.12 |
| 5 | Claude 2 | 66.46 |
| 6 | Gemini 1.0 Pro | 58.54 |
| 7 | deepseek-coder-1.3B-instruct | 57.32 |
| 8 | Phi-2 | 51.22 |
| 9 | starcoder2-15B | 46.34 |
| 10 | Mixtral 8x7B Instruct | 40.85 |
| 11 | starcoder2-7B | 37.8 |
| 12 | starcoder | 34.76 |
| 13 | gemma-7B | 33.54 |
| 14 | starcoder2-3B | 32.93 |
| 15 | Mistral 7B | 31.1 |
Interactive version: aibenchmarks.dev/benchmark?slug=evoeval-verbose · How the rankings work · Data refreshed daily, snapshot 2026-07-08.