RMGAP - Generative Verifier: 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; mean log-probability of the model answering YES to whether the response is good). Source: arxiv.org. Saturation forecast: Around December 2026. 2 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Qwen 3 8B | 67.13 |
| 2 | Qwen 2.5 7B Instruct | 66.81 |
Interactive version: theaggregate.ai/benchmark?slug=rmgap-generative-verifier · How It Works · Data refreshed daily, snapshot 2026-09-26.