← Arena leaderboards

Text · Style Control · Longer Query

Arena · 50 models

RankModelScore
1Gemini 4 Carbon (High)1,541.9336
2Gemini 4 Argon (High)1,538.9576
3Claude Opus 5.5 (High)1,525.2092
4Claude Fable 5 (High)1,522.1367
5Claude Opus 4.6 (High)1,521.4322
6Claude Opus 4.61,515.6745
7Claude Opus 4.7 (High)1,515.0139
8Claude Fable 5.1 (Max)1,511.8199
9Gemini 3.8 Flash (High)1,511.6404
10Claude Opus 4.71,507.4623
11Muse Wonder (Max)1,504.9617
12Claude Opus 5 (High)1,503.4543
13Claude Opus 4.8 (High)1,503.4339
14Kimi K3 (Max)1,502.9062
15Claude Opus 5 (Max)1,502.4357
16Gemini 3.7 Flash (High)1,501.0901
17Claude Opus 4.81,499.2568
18Claude Sonnet 5.5 (xHigh)1,498.9986
19Gemini 3.1 Pro Preview1,498.9693
20GPT-5.6 Sol (xHigh)1,497.5723
21Muse Spark 1.3 (Max)1,496.0068
22MiMo-V2.6-Pro1,494.9086
23Claude Opus 4.5 (High, 32K, 20251101)1,494.7845
24Claude Opus 4.5 (20251101)1,492.2093
25Claude Sonnet 4.61,492.083
26Qwen3.8-Max1,491.5199
27GPT-6.1 Sol (Max)1,491.1677
28Gemini 3 Pro1,490.2989
29Qwen3.7-Max-Preview1,490.1063
30Muse Spark 1.2 (xHigh)1,489.4048
31GPT-6 Astra (Max)1,488.0118
32GPT-5.5 (High)1,487.3636
33GPT-5.51,486.8656
34MiMo-V2.5-Pro1,486.0429
35GLM-5.3 (Max)1,485.9902
36GLM-5.2 (Max)1,485.6575
37Claude Sonnet 4.5 (High, 32K, 20250929)1,485.4898
38Gemini 3.6 Flash (High)1,485.2653
39Claude Opus 4.1 (Thinking 16K, 20250805)1,484.5023
40Gemini 3.5 Flash (High)1,482.3296
41GLM-5.3-Flash1,482.1844
42Claude Sonnet 4.5 (20250929)1,481.482
43Claude Sonnet 5 (High)1,481.4061
44Qwen3.5-Max-Preview1,481.0974
45GPT-5.4 (High)1,480.8559
46DeepSeek-V4.1-Flash-20260910 (Max)1,480.661
47Gemini 3.5 Flash (Medium)1,480.6535
48Muse Spark 1.11,480.4234
49GLM-5.11,480.0035
50Grok 4.51,479.6733