← Arena leaderboards

Text · Factuality

Arena · 50 models

RankModelScore
1Gemini 4 Carbon (High)1,521.5854
2Claude Opus 5.5 (High)1,508.6769
3Claude Fable 5.1 (Max)1,499.6873
4Gemini 4 Argon (High)1,499.5393
5Muse Wonder (Max)1,494.3282
6Claude Opus 4.6 (High)1,492.6206
7Claude Opus 5 (Max)1,488.3899
8Claude Opus 4.61,487.3636
9GPT-5.5 (High)1,485.6947
10MiMo-V2.6-Pro1,484.5005
11GPT-5.4 (High)1,484.2553
12Claude Fable 5 (High)1,482.7252
13GPT-5.51,481.8294
14Gemini 3 Pro1,481.2424
15Qwen3.7-Max-Preview1,479.6415
16Muse Spark 1.3 (Max)1,479.0976
17Muse Spark 1.2 (xHigh)1,479.0311
18Claude Opus 5 (High)1,478.2464
19Claude Opus 4.7 (High)1,477.3836
20Gemini 3.5 Flash (High)1,477.1554
21GPT-5.6 Sol (xHigh)1,476.3987
22Gemini 3.7 Flash (High)1,475.7613
23Claude Sonnet 5.5 (xHigh)1,475.5673
24Claude Opus 4.71,475.5438
25GPT-5.41,474.8512
26Qwen3.5-Max-Preview1,474.5884
27Gemini 3.8 Flash (High)1,471.9997
28Gemini 3.6 Flash (High)1,471.8834
29Gemini 3.1 Pro Preview1,471.6811
30Kimi K3 (Max)1,471.5128
31Qwen3.8-Max1,471.1131
32Claude Opus 4.8 (High)1,469.8135
33DeepSeek-V4.1-Flash-20260910 (Max)1,468.9852
34GPT-6 Luna (Max)1,468.7055
35GPT-5.6 Terra (xHigh)1,468.6812
36Muse Spark1,467.6315
37Gemini 3 Flash1,467.499
38Gemini 3.5 Flash (Medium)1,467.4855
39GPT-6 Astra (Max)1,466.5224
40GLM-5.3 (Max)1,466.112
41GPT-6.1 Sol (Max)1,465.5567
42Grok 4.51,465.1935
43Claude Opus 4.81,464.7281
44MiMo-V2.5-Pro1,464.3257
45GLM-5.3-Flash1,463.7881
46GLM-5.2 (Max)1,462.5993
47Muse Spark 1.11,462.4391
48Gemini 2.5 Pro1,461.7427
49Claude Sonnet 4.61,460.4823
50GPT-5.6 Luna (xHigh)1,460.4697