← Arena leaderboards

Text · Factuality · Math

Arena · 50 models

Rank	Model	Score
1	Claude Opus 5 (High)	1,516.5024
2	Claude Opus 5 (Max)	1,516.2227
3	Claude Fable 5.1 (Max)	1,512.8232
4	Claude Fable 5 (High)	1,508.1882
5	Claude Opus 4.6 (High)	1,506.2986
6	Gemini 3.8 Flash (High)	1,504.6315
7	Muse Spark 1.3 (Max)	1,499.0478
8	Gemini 3.7 Flash (High)	1,498.6988
9	GPT-5.5	1,495.2686
10	Claude Opus 4.6	1,493.6109
11	Qwen3.8-Max	1,493.2407
12	Gemini 3.5 Flash (High)	1,491.4891
13	GLM-5.3-Flash	1,490.864
14	GPT-5.4 (High)	1,490.2448
15	Claude Opus 4.7 (High)	1,489.9881
16	Gemini 3.6 Flash (High)	1,489.6701
17	GPT-5.5 (High)	1,487.0759
18	GPT-5.6 Sol (xHigh)	1,486.7128
19	Gemini 3.5 Flash (Medium)	1,484.5643
20	Inkling	1,481.8906
21	Claude Opus 4.8 (High)	1,480.6071
22	GLM-5.3 (Max)	1,478.7885
23	GPT-5.6 Terra (xHigh)	1,478.4754
24	Kimi K3 (Max)	1,478.468
25	Claude Opus 4.7	1,477.8461
26	MiMo-V2.5-Pro	1,477.6906
27	Gemini 3.1 Pro Preview	1,475.8749
28	Gemini 3 Pro	1,475.8651
29	Grok 4.5	1,474.2766
30	GPT-5.4	1,474.0676
31	Claude Sonnet 5 (High)	1,472.442
32	GLM-5.2 (Max)	1,472.0568
33	Kimi K2.6	1,471.7817
34	GPT-5.6 Luna (xHigh)	1,470.3856
35	ERNIE 5.1	1,468.4011
36	Muse Spark 1.1	1,468.3778
37	Gemini 3 Flash	1,467.1459
38	Claude Opus 4.8	1,466.5721
39	MiMo-V2.6-Flash	1,463.7072
40	Qwen3.7-Plus	1,463.1813
41	GLM-5.1	1,462.4304
42	Claude Opus 4.5 (High, 32K, 20251101)	1,460.685
43	Claude Sonnet 4.6	1,460.3586
44	Claude Opus 4.5 (20251101)	1,458.1458
45	Inkling-Small	1,457.9425
46	Qwen3.5-397B-A17B	1,455.7984
47	Kimi K2.5 (Thinking)	1,455.1322
48	Grok 4.20 Beta (Reasoning, 0309)	1,454.9196
49	GPT-5.2 (High)	1,454.2969
50	GLM-5	1,453.4294