← Arena leaderboards

Vision · Style Control · Entity Recognition

Arena · 50 models

RankModelScore
1Gemini 3 Pro1,291.657
2Gemini 3 Flash1,290.0107
3Gemini 3.1 Pro Preview1,283.9787
4Gemini 3 Flash (Minimal)1,273.9397
5Gemini 3.1 Flash-Lite Preview1,271.1111
6GPT-5 (High)1,261.4765
7Gemini 2.5 Pro1,250.8092
8Claude Opus 4.7 (High)1,242.3894
9GPT-5.2 Chat (20260210)1,238.8751
10Grok 4.20 Beta (Reasoning, 0309)1,238.1142
11Claude Opus 4.61,236.5651
12Grok 4 (0709)1,236.1557
13Claude Opus 4.6 (High)1,235.8476
14Kimi K2.5 (Thinking)1,232.3344
15o3 (2025-04-16)1,231.4361
16GPT-5.1 (High)1,228.2256
17Grok 4.20 Multi-Agent Beta (0309)1,226.5086
18Gemini 2.5 Flash1,221.3147
19GPT-5.5 (High)1,219.1786
20GPT-5.51,218.5757
21o4-mini (2025-04-16)1,218.556
22ChatGPT-4o Latest (20250326)1,218.1401
23Grok 4.1 Fast (Reasoning)1,216.6141
24GPT-5.41,215.376
25Qwen3.5-397B-A17B1,211.3363
26GPT-5 mini (High)1,211.1918
27Grok 4.31,209.5504
28GPT-5.2 (High)1,201.2996
29GPT-5.4 mini (High)1,198.1449
30GPT-4.1 mini (2025-04-14)1,196.7441
31Gemma 4 31B1,196.1957
32GPT-5.4 (High)1,196.1683
33GPT-5 Chat1,190.3673
34GPT-5.11,186.9041
35GPT-4.1 (2025-04-14)1,184.0948
36Claude Opus 4.71,183.9539
37Gemini 2.5 Flash-Lite Preview (Thinking, 06-17)1,183.371
38Qwen3-VL-235B-A22B-Instruct1,181.4717
39GLM-5V-Turbo1,174.6155
40GPT-5.21,174.5587
41Gemma 3 27B IT1,169.9598
42Claude Sonnet 4.61,166.5968
43MiMo-V2.51,162.5257
44MiniMax-M31,160.8475
45Gemma 4 26B A4B1,157.6845
46Qwen3.5-27B1,149.2257
47Mistral Small 3.2 (2506)1,134.3547
48Mistral Small 3.1 24B Instruct (2503)1,130.5295
49Mistral Medium 3.1 (2508)1,128.4312
50Mistral Medium 3 (2505)1,125.3955