Text · Factuality · German

Arena · 50 models

Rank	Model	Score
1	Gemini 3 Pro	1,476.8253
2	GPT-5.5	1,472.5228
3	Claude Opus 4.6 (High)	1,470.2423
4	Claude Opus 5 (High)	1,467.8446
5	GPT-5.4 (High)	1,466.9469
6	GPT-5.5 (High)	1,465.7404
7	Claude Opus 4.7 (High)	1,459.2457
8	Claude Opus 4.7	1,454.9345
9	Gemini 3.1 Pro Preview	1,453.0955
10	Claude Opus 4.6	1,451.9981
11	Gemini 2.5 Pro	1,447.5535
12	Claude Opus 4.8 (High)	1,446.7577
13	Claude Sonnet 5 (High)	1,446.6755
14	Gemini 3 Flash (Minimal)	1,442.8921
15	Inkling	1,442.59
16	DeepSeek-V4-Pro	1,441.5339
17	Claude Opus 4.8	1,437.5807
18	Grok 4.20 Multi-Agent Beta (0309)	1,436.0038
19	MiMo-V2.5-Pro	1,428.6772
20	Kimi K3 (Max)	1,423.7975
21	Claude Opus 4.5 (20251101)	1,423.3164
22	Grok 4.1	1,423.1639
23	GPT-5.4	1,422.3238
24	GPT-5.2	1,421.5554
25	GPT-5.1 (High)	1,419.7169
26	GLM-5.1	1,419.6861
27	Qwen3-235B-A22B-Instruct (2507)	1,414.608
28	Kimi K2.5 (Thinking)	1,414.2062
29	Qwen3.5-397B-A17B	1,414.1943
30	Gemini 2.5 Flash	1,413.2707
31	ChatGPT-4o Latest (20250326)	1,411.3532
32	Grok 4.20 Beta (Reasoning, 0309)	1,408.4733
33	Grok 4.1 (Thinking)	1,407.7629
34	MiniMax-M3	1,406.8107
35	Claude Sonnet 4.5 (20250929)	1,404.5484
36	Claude Sonnet 4.5 (High, 32K, 20250929)	1,402.937
37	Mistral Large 3	1,400.5294
38	Claude Sonnet 4.6	1,400.314
39	Mistral Medium 3.1 (2508)	1,399.4588
40	Gemini 3.1 Flash-Lite Preview	1,396.9384
41	GPT-5.4 nano (High)	1,394.3518
42	MiMo-V2-Flash (Non-Thinking)	1,389.8619
43	DeepSeek-V3.2	1,389.6437
44	Gemini 2.5 Flash-Lite Preview (Non-Thinking, 09-2025)	1,389.5351
45	Step-3.5-Flash	1,388.9995
46	MiniMax-M2.7	1,388.2844
47	GPT-6 Luna (Max)	1,386.9461
48	Grok 4.3	1,385.7604
49	Grok 4.1 Fast (Reasoning)	1,385.526
50	Kimi K2 Thinking Turbo	1,377.8626