AA LiveCodeBench — leaderboard
Artificial Analysis independent evaluation of LiveCodeBench: contamination-free coding from LeetCode, AtCoder, and CodeForces problems.
Metric: Pass@1 (%). Source: artificialanalysis.ai. Status: saturated. 343 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3 Pro (Preview) (High) | 91.75 |
| 2 | DeepSeek V3.2 Speciale | 89.63 |
| 3 | GPT-5.2 (Medium) | 89.42 |
| 4 | GLM-4.7 (Reasoning) | 89.42 |
| 5 | GPT-5.2 (xHigh) | 88.89 |
| 6 | GPT-OSS-120B (High) | 87.83 |
| 7 | Claude Opus 4.5 (Thinking) | 87.09 |
| 8 | GPT-5.1 (High) | 86.77 |
| 9 | MiMo-V2-Flash (Reasoning) | 86.77 |
| 10 | DeepSeek V3.2 (Thinking) | 86.24 |
| 11 | O4 Mini (High) | 85.93 |
| 12 | Kimi K2 (Thinking) | 85.29 |
| 13 | GPT-5.1 Codex (High) | 84.87 |
| 14 | GPT-5 (High) | 84.55 |
| 15 | GPT-5 Codex (High) | 84.02 |
Interactive version: theaggregate.ai/benchmark?slug=aa-livecodebench · How the rankings work · Data refreshed daily, snapshot 2026-07-22.