← Arena leaderboards

Vision · Style Control · Chinese

Arena · 50 models

RankModelScore
1Gemini 3.8 Flash (High)1,383.7345
2Gemini 3.7 Flash (High)1,368.6513
3Muse Spark 1.3 (xHigh)1,366.4903
4Claude Opus 5 (High)1,363.159
5GPT-6 Astra (Max)1,358.2119
6Claude Fable 5 (High)1,354.6073
7Claude Opus 4.6 (High)1,352.17
8Claude Fable 5.1 (Max)1,350.8198
9Qwen3.8-Max1,349.7901
10Claude Opus 4.61,347.8918
11Muse Spark 1.3 (Max)1,347.296
12Claude Opus 4.71,346.5692
13Gemini 3.5 Flash (High)1,346.4199
14Gemini 3.5 Flash (Medium)1,343.094
15Gemini 3 Pro1,342.667
16Gemini 3.1 Pro Preview1,339.1615
17GLM-5.3-Flash1,337.7538
18Claude Opus 4.7 (High)1,336.1517
19Muse Spark 1.11,335.4574
20Gemini 3.6 Flash (High)1,334.6185
21Claude Opus 4.81,334.0009
22Muse Spark1,333.5456
23Claude Opus 4.8 (High)1,330.7236
24Grok 4.51,330.1152
25GPT-5.51,329.3683
26Kimi K2.61,325.683
27GPT-5.5 (High)1,325.2044
28GPT-5.4 (High)1,316.4963
29Claude Sonnet 5 (High)1,316.2065
30Dola Seed 2.0 Pro1,313.8362
31Gemini 3 Flash1,313.5911
32GPT-5.6 Terra (xHigh)1,312.6114
33Qwen3.7-Plus1,312.1386
34MiMo-V2.51,311.6807
35DeepSeek-V4.1-Flash-20260910 (Max)1,311.414
36Kimi K2.5 (Instant)1,308.8497
37Gemini 3 Flash (Minimal)1,307.6928
38Gemma 4 31B1,305.452
39GPT-5.6 Sol (xHigh)1,305.1154
40Kimi K2.5 (Thinking)1,304.5636
41Gemini 3.5 Flash-Lite1,302.8361
42Claude Sonnet 4.61,301.6478
43GPT-5.2 (High)1,299.775
44GPT-5.41,297.3633
45Grok 4.20 Beta (Reasoning, 0309)1,297.2095
46GPT-5.5 Instant1,294.8569
47Qwen3.8-27B1,292.4479
48GPT-6 Luna (Max)1,291.6549
49Qwen3.5-397B-A17B1,291.2544
50Grok 4.20 Multi-Agent Beta (0309)1,291.1985