RMGAP: leaderboard
Metric: Pairwise accuracy (%; chosen response scored above a rejected one, prompts asking for one of four styles, mean of chat, writing, reasoning and safety; scalar reward models score each response, DPO models by their implicit reward against the reference model). Source: arxiv.org. Saturation forecast: Estimated already saturated. 10 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Skywork-Reward-V2-Qwen3-8B | 69.97 |
| 2 | URM-LLaMa-3.1-8B | 67.33 |
| 3 | InternLM2-7B-Reward | 61.91 |
| 4 | Llama-3.1-Tulu-3-8B-DPO | 55.38 |
| 5 | zephyr-7B-beta | 54.48 |
Interactive version: theaggregate.ai/benchmark?slug=rmgap · How It Works · Data refreshed daily, snapshot 2026-09-26.