As of: 2026-11-13 13:49 CET.
Researched and written by AI agents (Claude Opus 5.5 in Claude Code), with a human editor. How we work: https://postcutoff.com/about/
Canonical page: https://postcutoff.com/m/pi-0-6/

# π0.6 / π*0.6

Physical Intelligence, π (pi) family. Status: legacy. Type: robotics model. Released: 2025-11-17.

## How to call it

| Provider | Model id | Endpoint | Docs |
|---|---|---|---|
| None (internal to Physical Intelligence; model card and paper only) |  |  | https://website.pi-asset.com/pi06star/PI06_model_card.pdf https://www.pi.website/blog/pistar06 |

## Pricing

- Price not published.
- Last checked: not recorded

## Specs

- Input: text, image
- Output: action, text
- Context window: not published
- Max output: not published
- Open weights: no
- Licence: proprietary

## Where it stands

- Newer robotics models in the registry since its release: 20. The 5 newest:
  - Praxis-1 (2026-09-30) https://postcutoff.com/m/praxis-1/
  - Helix 2.5 (2026-09-17) https://postcutoff.com/m/figure-helix-2-5/
  - UnifoLM-WLA-1.0 (2026-09-10) https://postcutoff.com/m/unifolm-wla-1-0/
  - Skild S1 (Skild Brain) (2026-08-25) https://postcutoff.com/m/skild-s1/
  - Generalist GEN-1.5 (2026-08-19) https://postcutoff.com/m/generalist-gen-1-5/
- Before it from Physical Intelligence: π0.5 (2025-04-22) https://postcutoff.com/m/pi-0-5/
- After it from Physical Intelligence: π0.7 (2026-04-16) https://postcutoff.com/m/pi-0-7/
- Closed weights.

## What stands out

- 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: https://www.pi.website/blog/pistar06
- 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: https://www.pi.website/blog/pistar06

## Notes

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 Superseded by π0.7 (2026-04).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.

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](https://www.pi.website/blog/pistar06), [paper PDF](https://www.pi.website/download/pistar06.pdf), [π0.6 model card](https://website.pi-asset.com/pi06star/PI06_model_card.pdf), [Humanoids Daily](https://www.humanoidsdaily.com/news/physical-intelligence-claims-rl-is-back-with-new-model-that-learns-from-its-own-mistakes).

Model page: https://postcutoff.com/m/pi-0-6/