HELM Robo-Reward-Bench - Robo Reward Bench: leaderboard
Metric: EM. Source: crfm.stanford.edu. 14 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-4o (2024-11-20) | 65 |
| 2 | O1 (2024-12-17) | 63 |
| 3 | GPT-4.5 | 61 |
| 4 | Gemini 2.5 Flash (Preview 04-17) | 58.5 |
| 5 | Gemini 2.5 Flash (Preview 05-20) | 57 |
| 6 | Gemini 2.0 Flash Lite | 53.76 |
| 7 | Gemini 2.5 Pro (Preview 05-06) | 52.78 |
| 8 | Gemini 2.0 Flash | 48.37 |
Interactive version: theaggregate.ai/benchmark?slug=helm-robo-reward-bench-robo-reward-bench · How It Works · Data refreshed daily, snapshot 2026-09-08.