Vision Arena Overall
Multimodal · 2026-06-10
Arena-style leaderboard for multimodal AI models capable of reasoning over visual inputs, reported as arena/Elo-style model scores.
Top models (higher is better)
| Model | Score |
|---|---|
| Claude Fable 5 | 1309 |
| Qwen3.8-Max | 1300 |
| Opus 4.6 | 1299 |
| Opus 4.7 | 1299 |
| Gemini 3.7 Flash | 1295 |
| Muse Spark | 1294 |
| GPT-6.1 Sol | 1294 |
| Muse Spark 1.2 | 1293 |
| Gemini 3.8 Flash | 1291 |
| Muse Spark 1.3 | 1290 |
| Gemini 3 Pro Preview | 1289 |
| Claude Opus 5 | 1288 |
| Claude Fable 5.1 | 1288 |
| GPT-5.6 Sol | 1286 |
| GPT-5.5 | 1286 |
| Opus 4.8 | 1286 |
| GPT-6 Astra | 1285 |
| GPT-5.4 | 1284 |
| Gemini 3.5 Flash | 1282 |
| Gemini 3.6 Flash | 1282 |
| Muse Spark 1.1 | 1280 |
| Gemini 3.1 Pro Preview | 1280 |
| Grok 4.5 | 1279 |
| Sonnet 4.6 | 1275 |
| Gemini 3 Flash Preview | 1273 |
| GLM-5.3 Flash | 1271 |
| GPT-5.6 Terra | 1268 |
| Gemini 3.5 Flash-Lite | 1267 |
| Claude Sonnet 5.5 | 1265 |
| Kimi K2.6 | 1265 |
| Qwen3.7-Plus (source unspecified) | 1263 |
| Grok 4.6 | 1262 |
| Sonnet 5 | 1262 |
| Gemma 4 31B IT | 1262 |
| DeepSeek V4.1 Flash | 1261 |
| GPT-5.6 Luna | 1257 |
| Kimi K2.5 | 1252 |
| GPT-5.1 | 1251 |
| GPT-5.4 Mini | 1250 |
| Grok 4.7 | 1250 |