LMArena

Crowd preference Elo from blind head-to-head battles (formerly LMSYS Chatbot Arena).

#	Model	Maker	Score	Measured	Link
1	Claude Opus 5	Anthropic	1522	Oct 7, 2026	Source
2	GPT-5.6 Sol	OpenAI	1514	Oct 7, 2026	Source
3	Gemini 3.1 Pro Preview	Google DeepMind	1500	Oct 7, 2026	Source
4	Kimi K3	Moonshot AI	1500	Oct 7, 2026	Source
5	Claude Sonnet 5	Anthropic	1479	Oct 7, 2026	Source
6	DeepSeek V4 Pro (0813)	DeepSeek	1462	Oct 7, 2026	Source
7	Gemini 3.5 Flash-Lite	Google DeepMind	1418	Oct 14, 2026	Source
8	Llama 4 Maverick	Meta	1352	Oct 7, 2026	Source
9	Inkling	Thinking Machines Lab	1337	Oct 14, 2026	Source
10	Jamba Large 1.7	AI21 Labs	1197	Oct 14, 2026	Source

Scores are the benchmark owners’ published results, recorded with the date we verified them. This benchmark feeds the League Score — see methodology.