← Arena leaderboards

Text · Factuality · Exclude Ties

Arena · 50 models

RankModelScore
1Gemini 4 Carbon (High)1,547.6963
2Claude Opus 5.5 (High)1,526.6662
3Gemini 4 Argon (High)1,520.0847
4Claude Fable 5.1 (Max)1,517.6969
5Muse Wonder (Max)1,507.7335
6Claude Opus 4.6 (High)1,505.4846
7Claude Opus 5 (Max)1,502.3177
8Claude Opus 4.61,500.4516
9MiMo-V2.6-Pro1,497.7763
10Gemini 3 Pro1,493.7423
11GPT-5.5 (High)1,491.3628
12Claude Fable 5 (High)1,491.1424
13GPT-5.4 (High)1,490.3532
14Muse Spark 1.3 (Max)1,489.956
15Muse Spark 1.2 (xHigh)1,488.9418
16Claude Opus 5 (High)1,488.7778
17Qwen3.7-Max-Preview1,488.134
18GPT-5.51,487.3223
19Claude Opus 4.7 (High)1,486.5006
20Gemini 3.5 Flash (High)1,486.2026
21Qwen3.5-Max-Preview1,485.5823
22Gemini 3.7 Flash (High)1,485.293
23Claude Opus 4.71,484.2243
24Gemini 3.8 Flash (High)1,483.1421
25Claude Sonnet 5.5 (xHigh)1,482.6415
26Qwen3.8-Max1,480.5607
27GPT-5.6 Sol (xHigh)1,480.0485
28Gemini 3.6 Flash (High)1,479.6734
29Gemini 3.1 Pro Preview1,479.5127
30Kimi K3 (Max)1,478.977
31Gemini 3 Flash1,477.3272
32Muse Spark1,477.0451
33GPT-5.41,475.8986
34Gemini 3.5 Flash (Medium)1,475.5729
35Claude Opus 4.8 (High)1,474.3548
36DeepSeek-V4.1-Flash-20260910 (Max)1,474.2609
37GLM-5.3 (Max)1,473.1858
38Muse Spark 1.11,471.9496
39GPT-5.6 Terra (xHigh)1,470.3346
40GLM-5.3-Flash1,470.1379
41GLM-5.2 (Max)1,469.8252
42MiMo-V2.5-Pro1,469.0281
43Claude Opus 4.81,467.9077
44Grok 4.51,467.6205
45GPT-6 Astra (Max)1,466.6436
46GPT-6 Luna (Max)1,466.6431
47GPT-6.1 Sol (Max)1,466.4816
48Gemini 2.5 Pro1,466.3998
49Claude Sonnet 4.61,464.2823
50Claude Opus 4.5 (High, 32K, 20251101)1,462.736