LiveBench Story Generation — leaderboard

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

Top models

#ModelScore
1Gemini 1.5 Flash (Preview 0827)81.33
2Command-R+80.83
3Qwen 2 72B Instruct80.17
4GPT-4o (2024-08-06)80.08
5GPT-4o ChatGPT79.92
6GPT-4 Turbo79.92
7Mistral Large 2 (Jul)78.5
8DeepSeek V2.577.17
9Gemini 1.5 Pro (Preview 0801)77.08
10GPT-4 (0613)76.08
11Llama 3.1 405B Instruct75.5
12Llama 3.1 70B Instruct75.42
13DeepSeek Coder V275.33
14Gemini 1.5 Pro (Preview 0827)75
15Claude 3.5 Sonnet (20240620)74.33

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