Chatbot Arena (Text - Creative Writing): leaderboard

The arena.ai text board restricted to battles whose prompt asks for imaginative or artistic composition rather than factual reporting or analysis, a tag applied by a language model classifier. Voters compare two anonymous replies to one prompt and pick the better one. The rating is refit over this subset of votes alone, and higher is better.

Metric: Arena Score. Source: arena.ai. 412 models tracked.

Top models

#ModelScoreOverall rank
1Gemini 4 Argon (High)1519#8 (Gemini 4 Argon)
2Claude Opus 5.5 (High)1517#3 (Claude Opus 5.5)
3Claude Fable 5 (High)1502#16 (Claude Fable 5)
4Claude Opus 4.6 (High)1500#63 (Claude Opus 4.6)
5Gemini 3.7 Flash (High)1492#23 (Gemini 3.7 Flash)
6Claude Opus 4.7 (High)1489#49 (Claude Opus 4.7)
7Gemini 3.8 Flash (High)1487#21 (Gemini 3.8 Flash)
8Gemini 3 Pro1485#80
9Claude Opus 4.71483#49
10Claude Fable 5.1 (Max)1483#9 (Claude Fable 5.1)
11Gemini 3.1 Pro (Preview)1481#58
12Claude Opus 4.61479#63
13Claude Opus 5 (Max)1471#12 (Claude Opus 5)
14Gemini 3.6 Flash (High)1471#51 (Gemini 3.6 Flash)
15Claude Opus 5 (High)1471#12 (Claude Opus 5)

Interactive version: theaggregate.ai/benchmark?slug=chatbot-arena-text-creative-writing · How It Works · Data refreshed daily, snapshot 2026-10-20.