LMArena ELO

Overall

Human preference ranking from blind comparisons.

Metrics
ELO Score
Paper Dataset Eval Code

How to Run

Submit model to lmsys.org Arena for human evaluation (requires API endpoint)

Leaderboard

Rank Model Provider Parameters Score
1 Gemini 3 Pro Google Unknown 1501
2 Grok 4.1 xAI Unknown 1483
3 Claude Opus 4.5 Anthropic Unknown 1480
4 GPT-5.2 OpenAI Unknown 1475