LMArena Preference Proxy — leaderboard
Preference proxy evaluation from LMArena, measuring how well LLM evaluators reproduce preference signals across math, code, instruction-following, and knowledge tasks.
Metric: Evaluator Accuracy (%). Source: huggingface.co. 4 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemma 2 9B (IT) | 64.63 |
| 2 | Llama 3 8B Instruct | 64.01 |
| 3 | GPT-4o Mini (2024-07-18) | 58.83 |
| 4 | Claude 3 Haiku (20240307) | 57.39 |
Interactive version: theaggregate.ai/benchmark?slug=lmarena-preference-proxy · How the rankings work · Data refreshed daily, snapshot 2026-07-22.