← Arena leaderboards

Agent

Arena · 50 models

RankModelScore
1Claude Opus 5.5 (Max)0.2004
2Gemini 4 Carbon (High)0.1656
3Claude Sonnet 5.5 (Max)0.1462
4GPT-6 Astra (Max)0.1368
5Claude Fable 5.1 (Max)0.1315
6GPT-6.1 Sol (Max)0.1224
7Gemini 4 Argon (High)0.1061
8GPT-6.1 Terra (Max)0.1019
9Muse Wonder Full (Max)0.0963
10Claude Fable 5 (High)0.0778
11Claude Opus 5 (Max)0.076
12GPT-6 Sol (Max)0.0695
13Kimi K3.1 Pro (Max)0.047
14GPT-5.6 Sol (xHigh)0.0456
15Claude Sonnet 5 (High)0.0284
16Grok 4.7 (xHigh)0.0264
17Gemini 3.8 Flash (High)0.0256
18Muse Wonder (Max)0.0251
19Muse Spark 1.3 (Max)0.0244
20MiMo-V2.6-Pro0.0244
21DeepSeek-V4.1-Flash-20261017 (Max)0.0213
22GPT-6 Luna (Max)0.0192
23GLM-5.3 (Max)0.0117
24Hy4 preview0.0094
25Kimi K3 (Max)0.0086
26GLM-5.2 (Max)0.0075
27Claude Haiku 5.5 (High)0.0065
28MiMo-V2.6-Flash0.0043
29Qwen3.8-Max0.004
30Grok 4.6 (xHigh)0.0032
31DeepSeek-V4-Pro (High, 20260813)-0.003
32GPT-5.6 Terra (xHigh)-0.0151
33Gemini 3.7 Flash (High)-0.0163
34GPT-6 Luna (Medium)-0.0193
35GLM-5.3-Flash-0.0249
36Step 5 Preview-0.0253
37MiniMax-M3.1-Flash-0.0322
38Qwen3.8-27B-0.043
39Qwen3.8-Flash-Next-0.0433
40Gemini 3.1 Pro Preview-0.0461
41Beam 1.0 501B A23B-0.047
42GPT-6 Luna (Low)-0.0587
43Mistral Large 4-0.0641
44Qwen3.7-Max-0.0667
45Qwen3.7-Plus-0.0737
46Hy3-0.0753
47Muse Spark 1.1-0.0768
48Muse Spark 1.2 (xHigh)-0.0774
49MiMo-V2.5-Pro-0.0901
50Spark-X2.6-0.1