LabRobFail-Bench - Failure Classification: leaderboard

Metric: Top-1 accuracy (%; the failure category among five (perception, grasping, motion, logic, safety); robot-manipulation videos from simulated chemical self-driving laboratories with injected control, physics and semantic failures, seen-environment test split; answers from simulator ground truth; zero-shot VLM). Source: arxiv.org. Saturation forecast: Around 2029. 7 models tracked.

Top models

#ModelScore
1Gemini 2.5 Flash39.68
2GPT-5.438.54
3Gemini 2.0 Flash36.89

Interactive version: theaggregate.ai/benchmark?slug=labrobfail-bench-failure-classification · How It Works · Data refreshed daily, snapshot 2026-09-29.