Vision · Entity Recognition
Arena · 50 models
| Rank | Model | Score |
|---|---|---|
| 1 | Gemini 3 Pro | 1,347.5985 |
| 2 | Gemini 3 Flash | 1,335.5449 |
| 3 | Gemini 3.1 Pro Preview | 1,329.7023 |
| 4 | Gemini 3 Flash (Minimal) | 1,311.1244 |
| 5 | Gemini 3.1 Flash-Lite Preview | 1,307.593 |
| 6 | Grok 4 (0709) | 1,288.3946 |
| 7 | Kimi K2.5 (Thinking) | 1,284.8351 |
| 8 | Gemini 2.5 Pro | 1,283.1608 |
| 9 | Claude Opus 4.7 (High) | 1,277.8354 |
| 10 | Grok 4.20 Beta (Reasoning, 0309) | 1,277.0704 |
| 11 | Claude Opus 4.6 | 1,273.5532 |
| 12 | Claude Opus 4.6 (High) | 1,271.6666 |
| 13 | Grok 4.20 Multi-Agent Beta (0309) | 1,264.8204 |
| 14 | Qwen3.5-397B-A17B | 1,264.7777 |
| 15 | GPT-5 (High) | 1,262.581 |
| 16 | Gemini 2.5 Flash | 1,251.6437 |
| 17 | Grok 4.1 Fast (Reasoning) | 1,251.2585 |
| 18 | GPT-5.4 | 1,251.1294 |
| 19 | Qwen3-VL-235B-A22B-Instruct | 1,248.5405 |
| 20 | GPT-5.2 Chat (20260210) | 1,245.9786 |
| 21 | ChatGPT-4o Latest (20250326) | 1,245.0007 |
| 22 | GLM-5V-Turbo | 1,241.3476 |
| 23 | o3 (2025-04-16) | 1,233.8261 |
| 24 | GPT-5.2 (High) | 1,231.9641 |
| 25 | GPT-5.1 (High) | 1,231.8117 |
| 26 | GPT-5.5 | 1,231.2784 |
| 27 | GPT-5.4 (High) | 1,229.4049 |
| 28 | Claude Opus 4.7 | 1,226.1668 |
| 29 | GPT-5 mini (High) | 1,226.0322 |
| 30 | Gemma 4 31B | 1,225.7072 |
| 31 | GPT-5.5 (High) | 1,224.7201 |
| 32 | GPT-5 Chat | 1,219.081 |
| 33 | GPT-5.4 mini (High) | 1,218.2159 |
| 34 | o4-mini (2025-04-16) | 1,217.8753 |
| 35 | Grok 4.3 | 1,217.31 |
| 36 | Qwen3.5-27B | 1,210.1087 |
| 37 | GPT-4.1 (2025-04-14) | 1,207.5307 |
| 38 | GPT-5.2 | 1,197.4104 |
| 39 | MiniMax-M3 | 1,196.7495 |
| 40 | Gemma 3 27B IT | 1,196.0216 |
| 41 | Claude Sonnet 4.6 | 1,195.4717 |
| 42 | Gemma 4 26B A4B | 1,190.6465 |
| 43 | MiMo-V2.5 | 1,188.6438 |
| 44 | GPT-4.1 mini (2025-04-14) | 1,186.8495 |
| 45 | Gemini 2.5 Flash-Lite Preview (Thinking, 06-17) | 1,186.2979 |
| 46 | GPT-5.1 | 1,182.3583 |
| 47 | Mistral Small 3.1 24B Instruct (2503) | 1,154.7263 |
| 48 | Mistral Small 3.2 (2506) | 1,151.0638 |
| 49 | Mistral Medium 3.1 (2508) | 1,147.1319 |
| 50 | GPT-5.4 nano (High) | 1,140.7776 |