← Arena leaderboards

Text · Style Control · Expert

Arena · 50 models

RankModelScore
1Gemini 4 Carbon (High)1,581.385
2Claude Fable 5 (High)1,550.2791
3Claude Opus 4.6 (High)1,543.6586
4Gemini 4 Argon (High)1,541.2688
5Claude Opus 5.5 (High)1,537.6775
6Claude Opus 5 (High)1,537.3866
7Claude Opus 4.7 (High)1,533.9392
8Claude Opus 4.61,533.2697
9Claude Opus 4.71,533.1268
10Claude Fable 5.1 (Max)1,532.5817
11Claude Sonnet 5.5 (xHigh)1,532.2181
12Kimi K3 (Max)1,532.1563
13GPT-5.6 Sol (xHigh)1,531.7439
14MiMo-V2.6-Pro1,530.6893
15Claude Opus 5 (Max)1,529.8203
16Gemini 3.8 Flash (High)1,526.6601
17Claude Opus 4.8 (High)1,526.5024
18GPT-6.1 Sol (Max)1,525.6705
19Muse Spark 1.3 (Max)1,521.8763
20GLM-5.3 (Max)1,520.6466
21Gemini 3.7 Flash (High)1,517.7783
22GPT-6 Astra (Max)1,516.7132
23DeepSeek-V4.1-Flash-20260910 (Max)1,516.2585
24Muse Wonder (Max)1,516.1735
25Claude Opus 4.81,514.883
26GPT-5.4 (High)1,514.0988
27Qwen3.8-Max1,512.7268
28GLM-5.3-Flash1,512.52
29GPT-5.5 (High)1,512.031
30Claude Sonnet 5 (High)1,511.1376
31GPT-5.51,509.9171
32Gemini 3.1 Pro Preview1,509.2398
33Qwen3.7-Max-Preview1,506.9717
34GPT-5.6 Terra (xHigh)1,506.8977
35MiMo-V2.5-Pro1,505.7018
36Claude Sonnet 4.61,505.5248
37Claude Opus 4.5 (High, 32K, 20251101)1,503.002
38GLM-5.2 (Max)1,502.6191
39Muse Spark 1.11,502.5906
40Kimi K2.61,502.5387
41Claude Opus 4.5 (20251101)1,502.2143
42Gemini 3.6 Flash (High)1,500.0562
43Qwen3.6-Max-Preview1,499.6356
44Qwen3.5-Max-Preview1,499.1125
45Gemini 3 Pro1,498.7252
46Gemma 4 31B1,498.1465
47Claude Sonnet 4.5 (High, 32K, 20250929)1,497.6367
48Gemini 3 Flash1,497.1551
49GLM-5.11,496.6876
50GPT-6 Sol (Max)1,496.6715