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

#ModelScore
1Skywork-Reward-V2-Qwen3-8B69.97
2URM-LLaMa-3.1-8B67.33
3InternLM2-7B-Reward61.91
4Llama-3.1-Tulu-3-8B-DPO55.38
5zephyr-7B-beta54.48

Interactive version: theaggregate.ai/benchmark?slug=rmgap · How It Works · Data refreshed daily, snapshot 2026-09-26.