← Arena leaderboards

Text · Factuality · French

Arena · 50 models

RankModelScore
1Claude Opus 4.6 (High)1,529.7546
2Claude Opus 5 (High)1,526.7344
3GPT-5.4 (High)1,521.9844
4Claude Fable 5 (High)1,516.3914
5GPT-5.5 (High)1,515.9712
6Claude Opus 4.7 (High)1,514.6325
7Qwen3.8-Max1,514.1876
8GPT-5.51,513.179
9Gemini 3.6 Flash (High)1,511.9041
10GLM-5.3 (Max)1,509.0425
11Claude Opus 4.61,508.0807
12GPT-5.41,506.6477
13Gemini 3.5 Flash (High)1,506.4553
14GPT-5.6 Terra (xHigh)1,504.4354
15Claude Opus 4.81,503.8216
16Claude Opus 4.71,502.4637
17MiMo-V2.5-Pro1,502.3377
18Claude Opus 4.8 (High)1,501.8729
19Gemini 3.1 Pro Preview1,500.8815
20GPT-5.6 Sol (xHigh)1,500.5442
21Claude Opus 5 (Max)1,500.2593
22GPT-5.6 Luna (xHigh)1,499.6093
23Kimi K3 (Max)1,499.0945
24Claude Opus 4.5 (High, 32K, 20251101)1,498.2132
25GLM-5.3-Flash1,497.8226
26Gemini 3.8 Flash (High)1,494.25
27Grok 4.20 Beta (Reasoning, 0309)1,493.9254
28Kimi K2.61,490.7185
29DeepSeek-V4-Pro1,490.1027
30GPT-5.2 (High)1,489.2461
31Gemini 2.5 Pro1,488.9197
32Muse Spark 1.11,487.6448
33Claude Sonnet 4.61,487.105
34Gemini 3.5 Flash (Medium)1,487.0465
35Claude Sonnet 5 (High)1,486.5498
36ChatGPT-4o Latest (20250326)1,486.4439
37GPT-5.4 mini (High)1,484.1773
38GLM-5.2 (Max)1,483.1204
39Grok 4.20 Multi-Agent Beta (0309)1,482.4817
40GLM-5.11,481.4521
41GPT-5.1 (High)1,480.7778
42ERNIE 5.11,480.4982
43Qwen3.7-Plus1,480.1925
44GPT-5.21,478.2392
45Claude Opus 4.5 (20251101)1,478.0385
46Grok 4.51,476.8868
47Inkling1,476.8628
48Claude Sonnet 4.5 (20250929)1,476.7318
49Qwen3.5-397B-A17B1,475.5444
50Claude Sonnet 4.5 (High, 32K, 20250929)1,475.1704