← Arena leaderboards

Text · Factuality · Spanish

Arena · 50 models

Rank	Model	Score
1	Claude Opus 4.6	1,503.4943
2	Claude Opus 5 (Max)	1,491.0692
3	Claude Opus 4.6 (High)	1,489.0732
4	Claude Opus 4.7 (High)	1,486.6709
5	Gemini 3 Pro	1,483.6186
6	GPT-5.5 (High)	1,482.8723
7	Claude Fable 5 (High)	1,482.713
8	Gemini 3.8 Flash (High)	1,481.9076
9	Claude Opus 5 (High)	1,479.6176
10	Gemini 3.6 Flash (High)	1,479.3228
11	Gemini 3.5 Flash (Medium)	1,473.8126
12	Gemini 3.5 Flash (High)	1,473.4048
13	Gemini 3.1 Pro Preview	1,472.8865
14	Gemini 2.5 Pro	1,471.3741
15	Gemini 3.7 Flash (High)	1,470.9953
16	Claude Opus 4.8 (High)	1,470.8427
17	Claude Opus 4.7	1,470.6275
18	Qwen3.8-Max	1,470.5224
19	GLM-5.2 (Max)	1,470.2834
20	GPT-5.6 Sol (xHigh)	1,469.9831
21	GPT-5.6 Terra (xHigh)	1,469.9287
22	Kimi K3 (Max)	1,469.3176
23	Grok 4.5	1,468.4013
24	GPT-5.4	1,466.0517
25	MiMo-V2.5-Pro	1,463.7679
26	GPT-5.4 (High)	1,463.3535
27	GLM-5.3-Flash	1,460.5481
28	GPT-5.6 Luna (xHigh)	1,460.337
29	Inkling	1,458.5156
30	GPT-5.5	1,458.2935
31	Claude Sonnet 4.5 (20250929)	1,457.5072
32	Muse Spark 1.3 (Max)	1,457.3544
33	DeepSeek-V4-Pro	1,457.3194
34	ERNIE 5.1	1,457.3182
35	Claude Sonnet 4.6	1,457.2098
36	Claude Opus 4.8	1,456.9604
37	Qwen3.7-Plus	1,455.8005
38	Kimi K2.6	1,453.6257
39	Muse Spark 1.1	1,452.4795
40	GLM-5.1	1,452.29
41	Claude Opus 4.5 (High, 32K, 20251101)	1,451.8989
42	Claude Opus 4.5 (20251101)	1,450.5442
43	ERNIE 5.0 (0110)	1,448.9289
44	Grok 4.20 Multi-Agent Beta (0309)	1,448.6651
45	Qwen3.8-27B	1,448.1318
46	DeepSeek-V4-Pro Preview (High)	1,447.7105
47	Kimi K2.5 (Thinking)	1,445.6538
48	Claude Sonnet 4.5 (High, 32K, 20250929)	1,444.7017
49	Gemini 3.5 Flash-Lite	1,442.9324
50	GPT-5.1 (High)	1,441.6944