Leaderboard

Rank	Diff	Model	Score	95% CI	Votes	License
1	—	grok-4-fast xAI	1248		32	Proprietary
2	—	gemini-3-pro-preview Google	1169		28	Proprietary
3	—	gemini-3.1-pro-preview Google	1144		12	Proprietary
4	—	deepseek-v3.1-terminus DeepSeek	1141		39	Open Source
5	—	claude-sonnet-4.6 Anthropic	1130		21	Proprietary
6	—	claude-opus-4.1 Anthropic	1112		34	Proprietary
7	—	grok-4.1-fast xAI	1100		28	Proprietary
8	—	gpt-4.1 OpenAI	1091		35	Proprietary
9	—	qwen3-vl-235b-a22b-instruct Qwen	1070		24	Open Source
10	—	gpt-5.1 OpenAI	1067		25	Proprietary
11	—	grok-4 xAI	1062		33	Proprietary
12	—	kimi-k2-0905 Moonshot AI	1025		34	Open Source
13	—	gpt-5.2 OpenAI	1015		22	Proprietary
14	—	gpt-5-chat OpenAI	1006		35	Proprietary
15	—	o3 OpenAI	993		24	Proprietary
16	—	glm-4.6 Zhipu AI	988		26	Open Source
17	—	deepseek-r1-0528 DeepSeek	982		37	Open Source
18	—	qwen3-max Qwen	974		36	Proprietary
19	—	deepseek-chat-v3.1 DeepSeek	971		23	Open Source
20	—	qwen3-235b-a22b-2507 Qwen	967		30	Open Source
21	—	claude-opus-4 Anthropic	966		33	Proprietary
22	—	gemini-2.5-pro Google	915		31	Proprietary
23	—	claude-sonnet-4.5 Anthropic	913		37	Proprietary
24	—	gpt-4o OpenAI	909		12	Proprietary
25	—	gpt-5.3-chat OpenAI	908		11	Proprietary

Models are ranked by Bradley-Terry score on an Elo scale. 95% bootstrap confidence intervals show statistical uncertainty; once significance-aware ranking is enabled, models with overlapping intervals will share a tied rank.