← Arena leaderboards

Text · Style Control · Legal & Government

Arena · 50 models

RankModelScore
1Gemini 4 Carbon (High)1,549.7535
2Gemini 4 Argon (High)1,523.6179
3Muse Spark 1.2 (xHigh)1,518.8311
4Muse Wonder (Max)1,513.2944
5Claude Opus 4.7 (High)1,510.7629
6Claude Fable 5 (High)1,510.3317
7Kimi K3 (Max)1,509.1666
8Claude Opus 4.6 (High)1,509.1207
9Claude Opus 4.61,506.9938
10GPT-6.1 Sol (Max)1,504.8174
11Claude Opus 5 (Max)1,502.4396
12Claude Opus 5 (High)1,501.5387
13Gemini 3 Pro1,501.3708
14Muse Spark1,499.7459
15Claude Opus 5.5 (High)1,499.7289
16Muse Spark 1.3 (Max)1,499.4129
17Gemini 3.1 Pro Preview1,498.1603
18Muse Spark 1.11,497.8615
19Claude Opus 4.81,497.7278
20Gemini 3.7 Flash (High)1,495.6052
21Claude Opus 4.71,495.5827
22GPT-5.5 (High)1,495.5198
23Claude Opus 4.8 (High)1,495.4414
24GPT-5.6 Sol (xHigh)1,495.2911
25GLM-5.2 (Max)1,492.4508
26Gemini 3.8 Flash (High)1,492.3226
27Claude Fable 5.1 (Max)1,491.5857
28GPT-5.4 (High)1,490.1061
29Qwen3.8-Max1,489.1245
30Claude Opus 4.5 (High, 32K, 20251101)1,487.1738
31Claude Sonnet 5.5 (xHigh)1,486.2172
32GPT-5.51,486.0319
33Claude Sonnet 4.61,484.2934
34GPT-6 Astra (Max)1,484.1597
35Gemini 3.5 Flash (Medium)1,483.6758
36Gemini 3 Flash1,483.2517
37Claude Opus 4.5 (20251101)1,482.8035
38DeepSeek-V4-Pro (High, 20260813)1,480.8753
39GPT-5.2 Chat (20260210)1,480.7547
40Gemini 3.6 Flash (High)1,480.389
41Qwen3.7-Max-Preview1,479.7874
42Gemini 3.5 Flash (High)1,478.6947
43Grok 4.20 Beta (Reasoning, 0309)1,478.2576
44Grok 4.51,477.3923
45GPT-5.5 Instant1,476.4346
46GPT-5.41,475.276
47Grok 4.1 (Thinking)1,475.0883
48MiMo-V2.5-Pro1,474.9859
49GPT-5.6 Terra (xHigh)1,474.877
50DeepSeek-V4-Pro1,474.4406