RoboBench — leaderboard
Embodied-brain benchmark for multimodal LLMs across perception, instruction comprehension, planning, affordance prediction, and failure analysis.
Metric: Overall Dimension Average (self-reported). Source: benchmarklist.com. Status: saturation imminent. 15 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 2.5 Pro | 50.1 |
| 2 | Gemini 2.5 Flash | 45.06 |
| 3 | Gemini 2.0 Flash | 45.04 |
| 4 | Claude 3.7 Sonnet | 40.53 |
| 5 | GPT-4o | 40.16 |
| 6 | Claude 3.5 Sonnet | 37.82 |
| 7 | GPT-4o Mini | 34.4 |
Interactive version: theaggregate.ai/benchmark?slug=robobench · How the rankings work · Data refreshed daily, snapshot 2026-07-22.