Chatbot Arena (Vision - Creative Writing): leaderboard
The arena.ai vision board restricted to image battles that ask for writing outside academic, professional and technical registers, calling for imagination or storytelling. Voters compare two anonymous answers to the same image and prompt. The rating is refit over the conversations in this category alone, and higher is better.
Metric: Arena Score. Source: arena.ai. 101 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3 Pro | 1325 | #80 |
| 2 | Claude Fable 5 (High) | 1320 | #16 (Claude Fable 5) |
| 3 | Claude Opus 4.6 (High) | 1309 | #63 (Claude Opus 4.6) |
| 4 | Claude Opus 4.7 (High) | 1302 | #49 (Claude Opus 4.7) |
| 5 | Claude Opus 5.5 (High) | 1301 | #3 (Claude Opus 5.5) |
| 6 | Gemini 3.1 Pro (Preview) | 1300 | #58 |
| 7 | Muse Spark | 1300 | #64 |
| 8 | Claude Opus 4.6 | 1298 | #63 |
| 9 | Gemini 3 Flash | 1297 | #99 |
| 10 | Claude Opus 4.7 | 1297 | #49 |
| 11 | Gemini 3.7 Flash (High) | 1293 | #23 (Gemini 3.7 Flash) |
| 12 | GPT-5.6 Sol (xHigh) | 1292 | #19 (GPT-5.6 Sol) |
| 13 | Qwen 3.8 Max | 1290 | #35 |
| 14 | Claude Fable 5.1 (Max) | 1290 | #9 (Claude Fable 5.1) |
| 15 | GPT-6 Astra (Max) | 1289 | #6 (GPT-6 Astra) |
Interactive version: theaggregate.ai/benchmark?slug=chatbot-arena-vision-creative-writing · How It Works · Data refreshed daily, snapshot 2026-10-20.