← Arena leaderboards

Text · Factuality · Multi Turn

Arena · 50 models

RankModelScore
1Gemini 4 Argon (High)1,524.8876
2Claude Opus 5.5 (High)1,509.4995
3Claude Opus 4.6 (High)1,507.6468
4Muse Wonder (Max)1,506.2068
5Claude Fable 5 (High)1,503.02
6Claude Opus 4.61,502.1629
7GPT-5.4 (High)1,499.9093
8Claude Opus 4.7 (High)1,497.9023
9Claude Opus 5 (Max)1,497.0234
10Claude Opus 4.71,495.9592
11Gemini 3.8 Flash (High)1,492.9413
12GPT-5.5 (High)1,492.5988
13Gemini 3 Pro1,492.3041
14Kimi K3 (Max)1,491.8829
15GPT-5.51,491.2113
16Claude Fable 5.1 (Max)1,490.9341
17Muse Spark 1.3 (Max)1,490.5609
18Muse Spark 1.2 (xHigh)1,490.3808
19Claude Opus 4.8 (High)1,490.0948
20Gemini 3.5 Flash (High)1,489.9025
21Qwen3.8-Max1,489.2756
22GPT-5.41,488.3842
23Claude Opus 5 (High)1,487.5397
24Qwen3.5-Max-Preview1,487.3353
25Gemini 3.7 Flash (High)1,484.8705
26Claude Opus 4.81,483.9767
27Gemini 3.1 Pro Preview1,482.9501
28Gemini 3.6 Flash (High)1,481.8923
29GPT-5.6 Sol (xHigh)1,481.7198
30GLM-5.3 (Max)1,481.1998
31MiMo-V2.5-Pro1,480.6189
32GPT-5.6 Terra (xHigh)1,480.2219
33Muse Spark 1.11,479.2158
34Claude Opus 4.5 (20251101)1,477.9144
35Muse Spark1,477.6192
36Gemini 3.5 Flash (Medium)1,476.9104
37GPT-6 Astra (Max)1,476.0225
38GLM-5.11,475.9008
39DeepSeek-V4-Pro1,475.8446
40Grok 4.51,475.5798
41Claude Opus 4.5 (High, 32K, 20251101)1,475.4519
42GLM-5.2 (Max)1,474.8979
43DeepSeek-V4.1-Flash-20260910 (Max)1,473.6836
44Claude Sonnet 5.5 (xHigh)1,472.5776
45Claude Sonnet 4.5 (20250929)1,471.9597
46Gemini 3 Flash1,471.5848
47MiMo-V2.6-Pro1,471.4913
48Claude Sonnet 4.61,470.0617
49GLM-5.3-Flash1,469.7268
50GPT-5.1 (High)1,469.049