preference
LMSYS Chatbot Arena
Blind pairwise human preference voting on real-world chat prompts. Rankings computed via Bradley-Terry. The dominant industry preference signal for chat quality.
Official leaderboard →Top scores
| # | Model | Score |
|---|---|---|
| 1 | gemini-2-5-pro | 1421 Elo |
| 2 | gpt-5 | 1418 Elo |
| 3 | claude-opus-4-8 | 1397 Elo |
| 4 | grok-4 | 1381 Elo |
| 5 | deepseek-v3 | 1370 Elo |
Scores are snapshots from public leaderboards at the time of last update. Follow the source link for the live board.