LiveBench Olympiad — leaderboard
Metric: Score. Source: livebench.ai. 73 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 1.5 Pro (Preview 0827) | 68.66 |
| 2 | Claude 3.5 Sonnet (20240620) | 67.46 |
| 3 | GPT-4o (2024-08-06) | 65.96 |
| 4 | GPT-4o (2024-05-13) | 65.06 |
| 5 | DeepSeek Coder V2 | 64.61 |
| 6 | Mistral Large 2 (Jul) | 64.41 |
| 7 | DeepSeek V2.5 | 64.29 |
| 8 | Llama 3.1 405B Instruct | 64.15 |
| 9 | GPT-4o ChatGPT | 63.78 |
| 10 | Gemini 1.5 Flash (Preview 0827) | 63.4 |
| 11 | Gemini 1.5 Pro (Preview 0801) | 61.88 |
| 12 | Gemini 1.5 Flash (0514) | 61.67 |
| 13 | GPT-4 Turbo | 60.19 |
| 14 | Llama 3.1 70B Instruct | 60.06 |
| 15 | Claude 3 Opus (20240229) | 59.96 |
Interactive version: theaggregate.ai/benchmark?slug=livebench-olympiad · How the rankings work · Data refreshed daily, snapshot 2026-07-22.