← Arena leaderboards

Text · Style Control · Exclude Ties

Arena · 50 models

RankModelScore
1Gemini 4 Carbon (High)1,554.1923
2Gemini 4 Argon (High)1,546.9323
3Muse Wonder (Max)1,530.0533
4Claude Opus 5.5 (High)1,529.5844
5Claude Fable 5.1 (Max)1,524.0517
6Claude Fable 5 (High)1,523.2149
7Claude Opus 4.6 (High)1,522.0804
8Claude Opus 4.7 (High)1,518.7229
9Claude Opus 4.61,514.6451
10Gemini 3.8 Flash (High)1,513.8537
11Muse Spark 1.3 (Max)1,512.7802
12Muse Spark 1.2 (xHigh)1,508.3685
13Claude Opus 4.71,508.153
14Claude Opus 5 (High)1,506.7054
15Muse Spark 1.11,505.7441
16Claude Opus 5 (Max)1,505.22
17Gemini 3.7 Flash (High)1,504.885
18Kimi K3 (Max)1,502.4451
19Muse Spark1,502.137
20Gemini 3.1 Pro Preview1,501.5862
21Gemini 3 Pro1,501.1142
22GPT-5.6 Sol (xHigh)1,500.4856
23Gemini 3.6 Flash (High)1,495.5574
24Qwen3.8-Max1,495.4152
25GPT-6.1 Sol (Max)1,494.9149
26Claude Opus 4.8 (High)1,494.0363
27GPT-5.5 (High)1,492.3097
28GLM-5.3 (Max)1,489.8429
29MiMo-V2.6-Pro1,489.5365
30GPT-6 Astra (Max)1,487.8347
31Gemini 3.5 Flash (High)1,487.618
32Gemini 3.5 Flash (Medium)1,487.3314
33Claude Sonnet 5.5 (xHigh)1,486.9424
34GPT-5.2 Chat (20260210)1,486.3337
35GPT-5.51,486.0441
36DeepSeek-V4.1-Flash-20260910 (Max)1,485.6463
37GLM-5.3-Flash1,485.1155
38Claude Opus 4.5 (High, 32K, 20251101)1,484.9856
39Gemini 3 Flash1,484.9847
40GLM-5.2 (Max)1,484.9679
41GPT-5.4 (High)1,484.445
42Claude Opus 4.81,484.4331
43GPT-5.5 Instant1,484.3961
44Grok 4.20 Beta 11,484.029
45Claude Sonnet 4.61,481.558
46Qwen3.7-Max-Preview1,481.3138
47Grok 4.20 Beta (Reasoning, 0309)1,479.8037
48Claude Opus 4.5 (20251101)1,478.6481
49Grok 4.20 Multi-Agent Beta (0309)1,477.7454
50GPT-5.6 Terra (xHigh)1,475.1275