Chatbot Arena (Vision - English): leaderboard
The arena.ai vision board restricted to image battles whose prompt is in English. Each battle pairs an uploaded image and a prompt with two anonymous model answers, and the voter picks the better one. The rating is refit over the conversations in this category alone, so the order can differ from the overall vision table, and higher is better.
Metric: Arena Score. Source: arena.ai. 165 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Fable 5 (High) | 1305 | #16 (Claude Fable 5) |
| 2 | Claude Opus 4.7 | 1302 | #49 |
| 3 | Claude Opus 4.7 (High) | 1302 | #49 (Claude Opus 4.7) |
| 4 | Claude Opus 4.6 (High) | 1301 | #63 (Claude Opus 4.6) |
| 5 | Claude Opus 4.6 | 1296 | #63 |
| 6 | Qwen 3.8 Max | 1294 | #35 |
| 7 | Muse Spark | 1290 | #64 |
| 8 | GPT-6 Astra (Max) | 1288 | #6 (GPT-6 Astra) |
| 9 | GPT-6.1 Sol (Max) | 1287 | #7 (GPT-6.1 Sol) |
| 10 | Gemini 3.8 Flash (High) | 1286 | #21 (Gemini 3.8 Flash) |
| 11 | Gemini 3.7 Flash (High) | 1286 | #23 (Gemini 3.7 Flash) |
| 12 | Claude Opus 5.5 (High) | 1286 | #3 (Claude Opus 5.5) |
| 13 | GPT-5.6 Sol (xHigh) | 1284 | #19 (GPT-5.6 Sol) |
| 14 | Claude Sonnet 4.6 | 1282 | #90 |
| 15 | Claude Opus 4.8 | 1282 | #37 |
Interactive version: theaggregate.ai/benchmark?slug=chatbot-arena-vision-english · How It Works · Data refreshed daily, snapshot 2026-10-20.