π0.6 / π*0.6
π0.6: ~5B-parameter VLA with a Gemma 3 4B backbone and ~860M-parameter action expert, keeps π0.5's hierarchical design (per model card, 2025-11-17, via search snippet). No weights or API. Superseded by π0.7 (2026-04). pi.website blocked automated fetches on 2026-09-29; details taken from search snippets of the blog/model card.
- Input
- text, image
- Output
- action, text
- License
- proprietary
How to call it
| Provider | Model id | Endpoint / URL | Docs |
|---|---|---|---|
| None (internal to Physical Intelligence; model card and paper only) | — | www.pi.website/blog/pistar06 | docs |
Notable capabilities (2)
- Recap - RL from real-world experience and corrections: π*0.6 improves π0.6 with Recap (RL with Experience & Corrections via Advantage-conditioned Policies): demonstrations, then human interventions, then autonomous-trial RL; over 2x throughput and roughly halved failure rates on hard tasks. source
- Hours-long autonomous operation: Made espresso drinks for 18 hours straight, folded 50 novel laundry items in a new home, and assembled/labeled 59 factory boxes. source
Changelog: 2026-09-29 linked new entry 2025-11-17-physical-intelligence-pi-star-0-6-recap and arXiv 2511.14759.
Sources: π*0.6 blog, paper PDF, π0.6 model card, Humanoids Daily.
Timeline entry
- Physical Intelligence's π*0.6 learns from real-world experience with RL (Recap), running tasks for hours ★★★
On 2025-11-17 Physical Intelligence released π*0.6, a version of its π0.6 VLA improved with Recap (RL with Experience & Corrections via Advantage-conditioned Policies): demonstrations, then human corrections, then RL on the robot's own autonomous trials. Recap more than doubled throughput and…
Other Physical Intelligence models
π0.7 · π0.5 · π0-FAST · π0 (pi-zero)