← Arena leaderboards

Text · Instruction Following

Arena · 50 models

RankModelScore
1Gemini 4 Carbon (High)1,539.1879
2Claude Opus 5.5 (High)1,538.9173
3Gemini 4 Argon (High)1,532.5704
4Claude Opus 4.6 (High)1,520.8533
5Claude Fable 5.1 (Max)1,518.6833
6Claude Opus 5 (Max)1,518.0362
7Claude Opus 5 (High)1,515.3222
8Claude Opus 4.61,509.2018
9Claude Fable 5 (High)1,502.5171
10Claude Opus 4.7 (High)1,497.8677
11Muse Wonder (Max)1,493.6175
12Gemini 3.8 Flash (High)1,492.6746
13Claude Sonnet 5.5 (xHigh)1,491.8396
14MiMo-V2.6-Pro1,491.6718
15Claude Opus 4.71,488.4747
16Gemini 3.7 Flash (High)1,485.7854
17Kimi K3 (Max)1,485.2477
18GPT-5.6 Sol (xHigh)1,482.7621
19GLM-5.3 (Max)1,481.2641
20GLM-5.3-Flash1,480.6077
21Qwen3.8-Max1,480.4341
22GPT-5.5 (High)1,478.5634
23Claude Opus 4.5 (High, 32K, 20251101)1,478.0269
24Muse Spark 1.3 (Max)1,477.6078
25Claude Opus 4.8 (High)1,476.9736
26Claude Sonnet 4.61,475.7135
27MiMo-V2.5-Pro1,475.4406
28GPT-5.51,475.1328
29DeepSeek-V4.1-Flash-20260910 (Max)1,473.7945
30Claude Opus 4.5 (20251101)1,472.3706
31GPT-5.4 (High)1,469.3061
32GPT-6.1 Sol (Max)1,467.3614
33Gemini 3.5 Flash (High)1,467.1933
34Gemini 3.6 Flash (High)1,467.028
35Qwen3.5-Max-Preview1,466.813
36Gemini 3.1 Pro Preview1,466.0365
37GLM-5.2 (Max)1,465.6056
38Gemini 3.5 Flash (Medium)1,464.9186
39Muse Spark 1.2 (xHigh)1,464.8438
40Claude Opus 4.81,462.9957
41ERNIE 5.11,460.3263
42Qwen3.7-Max-Preview1,459.9896
43Claude Sonnet 4.5 (20250929)1,459.1552
44Claude Sonnet 4.5 (High, 32K, 20250929)1,458.0317
45Gemini 3 Pro1,457.8406
46Muse Spark 1.11,456.7262
47GPT-5.41,454.9351
48GPT-5.6 Terra (xHigh)1,454.5142
49GPT-6 Astra (Max)1,454.062
50Claude Sonnet 5 (High)1,452.6828