CodeElo — leaderboard
Metric: Elo Rating. Source: codeelo-bench.github.io. 33 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Human Expert | 3979 |
| 2 | O1 Mini | 1578 |
| 3 | QwQ 32B-Preview | 1261 |
| 4 | Median Human | 1200 |
| 5 | Qwen 2.5 72B Instruct | 634 |
| 6 | Mistral Large 2 (Nov) Instruct (2411) | 631 |
| 7 | Qwen 2.5 Coder 32B Instruct | 575 |
| 8 | Qwen 2.5 32B Instruct | 513 |
| 9 | Llama 3.1 70B Instruct | 478 |
| 10 | Qwen 2.5 Coder 14B Instruct | 424 |
| 11 | Qwen 2.5 14B Instruct | 414 |
| 12 | Qwen 2.5 Coder 7B Instruct | 397 |
| 13 | Codestral-22B-v0.1 | 385 |
| 14 | Qwen 2.5 7B Instruct | 315 |
| 15 | Yi-Coder-9B-Chat | 296 |
Interactive version: theaggregate.ai/benchmark?slug=codeelo · How the rankings work · Data refreshed daily, snapshot 2026-07-22.