Benchmarks · other
Arena Vision · English
Human preference on prompts with image inputs (multimodal understanding); Bradley-Terry rating. Only the English category.
107 models · 116 results · updated 2026-11-09 · arena.ai ↗
Leaderboard
Fable 5 high
All results
116 rows| # | Model | Effort | Arena Vision · English |
|---|---|---|---|
| 1 | Fable 5 | high | 1306 ±10 |
| 2 | Opus 4.7 | max | 1302 ±9 |
| 3 | Opus 4.7 | high | 1302 ±9 |
| 4 | Opus 4.6 | high | 1302 ±9 |
| 5 | Opus 4.6 | max | 1295 ±8 |
| 6 | Qwen 3.8 Max | max | 1294 ±11 |
| 7 | Muse Spark | 1291 ±13 | |
| 8 | Gemini 3.8 Flash | high | 1290 ±13 |
| 9 | GPT-6 Astra | max | 1289 ±15 |
| 10 | Opus 5.5 | high | 1289 ±17 |
| 11 | GPT-5.6 Sol | xhigh | 1288 ±11 |
| 12 | Gemini 3.7 Flash | high | 1287 ±14 |
| 13 | GPT-6.1 Sol | max | 1287 ±18 |
| 14 | Fable 5.1 | max | 1283 ±14 |
| 15 | Opus 4.8 | max | 1282 ±9 |
| 16 | GPT-5.5 | high | 1282 ±8 |
| 17 | Sonnet 4.6 | max | 1282 ±8 |
| 18 | Muse Spark 1.3 | max | 1282 ±13 |
| 19 | GPT-5.5 | xhigh | 1281 ±8 |
| 20 | GPT-5.2 | xhigh | 1280 ±9 |
| 21 | Opus 4.8 | high | 1280 ±9 |
| 22 | GPT-5.4 | high | 1279 ±8 |
| 23 | GPT-5.5 Instant | 1279 ±12 | |
| 24 | Muse Spark 1.3 | xhigh | 1278 ±13 |
| 25 | Gemini 3.5 Flash | high | 1277 ±10 |
| 26 | Opus 5 | high | 1277 ±9 |
| 27 | GPT-5.4 | xhigh | 1277 ±9 |
| 28 | Muse Spark 1.2 | xhigh | 1276 ±24 |
| 29 | Muse Spark 1.1 | max | 1276 ±10 |
| 30 | Gemini 3 Pro | high | 1275 ±10 |