LiveBench Table Reformat — leaderboard

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

Top models

#ModelScore
1GPT-4 Preview (0125)70
2Llama 3.1 70B Instruct68
3Claude 3.5 Sonnet (20240620)68
4Gemini 1.5 Pro (Preview 0801)68
5Gemini 1.5 Pro (Preview 0827)68
6Hermes 3 - Llama-3.1 70B66
7Gemini 1.5 Pro (0514)66
8Gemma 2 27B (IT)64
9GPT-4o (2024-05-13)64
10GPT-4 Turbo64
11Llama 3.1 405B Instruct64
12Mistral Large 2 (Jul)64
13Claude 3 Opus (20240229)64
14Mistral Large64
15DeepSeek V2.564

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