Vision · Captioning
Arena · 34 models
| Rank | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Pro Preview | 1,344.8858 |
| 2 | Gemini 3 Pro | 1,325.3215 |
| 3 | Gemini 2.5 Pro | 1,303.5664 |
| 4 | Gemma 4 26B A4B | 1,296.7088 |
| 5 | Qwen3-VL-235B-A22B-Instruct | 1,281.3241 |
| 6 | Gemini 3 Flash | 1,281.3181 |
| 7 | Gemini 2.5 Flash | 1,276.646 |
| 8 | GPT-5.2 (High) | 1,266.637 |
| 9 | GLM-5V-Turbo | 1,264.1626 |
| 10 | Gemini 3.1 Flash-Lite Preview | 1,259.4524 |
| 11 | Kimi K2.5 (Thinking) | 1,257.1273 |
| 12 | GPT-5.1 (High) | 1,251.6128 |
| 13 | Grok 4 (0709) | 1,245.3309 |
| 14 | Gemini 3 Flash (Minimal) | 1,240.5873 |
| 15 | Qwen3.5-397B-A17B | 1,229.6993 |
| 16 | ChatGPT-4o Latest (20250326) | 1,223.7982 |
| 17 | Gemma 4 31B | 1,217.5249 |
| 18 | GPT-5 mini (High) | 1,216.962 |
| 19 | GPT-5 Chat | 1,211.5692 |
| 20 | o3 (2025-04-16) | 1,209.741 |
| 21 | GPT-4.1 (2025-04-14) | 1,206.7569 |
| 22 | o4-mini (2025-04-16) | 1,204.0381 |
| 23 | GPT-5 (High) | 1,203.7575 |
| 24 | Gemini 2.5 Flash-Lite Preview (Thinking, 06-17) | 1,184.0261 |
| 25 | GPT-4.1 mini (2025-04-14) | 1,173.9566 |
| 26 | GPT-5.2 | 1,173.7762 |
| 27 | Qwen3.5-27B | 1,170.2042 |
| 28 | GPT-5.1 | 1,165.3876 |
| 29 | Mistral Small 3.1 24B Instruct (2503) | 1,152.8906 |
| 30 | Gemma 3 27B IT | 1,151.1786 |
| 31 | Mistral Medium 3.1 (2508) | 1,144.6216 |
| 32 | Gemini 2.0 Flash (001) | 1,129.5902 |
| 33 | Mistral Medium 3 (2505) | 1,094.2471 |
| 34 | Mistral Small 3.2 (2506) | 1,058.9403 |