Fine-tuning π0.5 on a real factory task: data diversity beats raw scale, hitting 98% success

DominiqueCAPaul · x · 2026-09-24

Dream Machines' first blog post documents two months of ablation experiments fine-tuning Physical Intelligence's open-source VLA model π0.5 on a real manufacturing task from a German manufacturer, reaching 98% policy success rate. All results, data, and runs—including failures—are being published.

The task requires two arms to transfer actuators from a box into a fixture with correct orientation. The authors note that neither the π0.5 paper nor the official repo offers fine-tuning recommendations, and no other company has published results, so these findings aim to save others time on tuning and data collection.

Related event: Real-World π0.5 Fine-Tuning: Data Diversity Beats Scale, Hitting 98% Success(6 posts)→

Original post →

More from Embodied

Embodied channel →