LiveBench Olympiad — leaderboard

Metric: Score. Source: livebench.ai. 73 models tracked.

Top models

#ModelScore
1Gemini 1.5 Pro (Preview 0827)68.66
2Claude 3.5 Sonnet (20240620)67.46
3GPT-4o (2024-08-06)65.96
4GPT-4o (2024-05-13)65.06
5DeepSeek Coder V264.61
6Mistral Large 2 (Jul)64.41
7DeepSeek V2.564.29
8Llama 3.1 405B Instruct64.15
9GPT-4o ChatGPT63.78
10Gemini 1.5 Flash (Preview 0827)63.4
11Gemini 1.5 Pro (Preview 0801)61.88
12Gemini 1.5 Flash (0514)61.67
13GPT-4 Turbo60.19
14Llama 3.1 70B Instruct60.06
15Claude 3 Opus (20240229)59.96

Interactive version: theaggregate.ai/benchmark?slug=livebench-olympiad · How the rankings work · Data refreshed daily, snapshot 2026-07-22.