Post-Cutoff

Physical Intelligenceπ (pi) familyCurrentRobotics modelReleased 22 Apr 2025

π0.5

Model idgs://openpi-assets/checkpoints/pi05_base

PricePrice not published

Checked29 September 2026

Where it stands

26newer robotics models since its release

The 5 newest are listed. All robotics models

Before it from Physical Intelligence
π0-FAST16 Jan 2025
After it from Physical Intelligence
π0.6 / π*0.617 Nov 2025
Weights
Weights are public. Licence: apache-2.0; weights built on PaliGemma, Gemma terms apply. See the licence for commercial terms.

How to call it

2 ways to use π0.5, with the model id each provider expects.

Providers, model ids, endpoints and docs for π0.5
ProviderModel idEndpointDocs
GitHub (openpi, JAX + PyTorch) gs://openpi-assets/checkpoints/pi05_base github.com
Hugging Face (LeRobot port) lerobot/pi05_base Docs
Model page

Pricing

Price not published.

No price source linked. Checked 29 September 2026.

Specs

Input
Text and images
Output
Robot actions and text
Context window
Not published
Max output
Not published
Open weights
Yes
Licence
apache-2.0 (openpi code); weights built on PaliGemma, Gemma terms apply
Released
22 April 2025
Status
Current
Type
Robotics model

What stands out

What its maker marketed at launch, and what people found later, each with its source.

  • Open-world generalization to unseen homes

    Cleans kitchens and bedrooms in entirely new homes not in training; performance improved as training grew from 3 to 104 homes; co-trained on heterogeneous robot, web and verbal-instruction data.

    Source: pi.website

  • Hierarchical subtask prediction + actions in one model

    Predicts a high-level text subtask, then low-level actions; trained with knowledge insulation.

    Source: github.com

  • Newest open-weights π model

    Base plus LIBERO and DROID checkpoints released in openpi in September 2025; the latest PI model with public weights as of 2026-09 (π0.6/π0.7 are closed).

    Source: github.com

Notes

Announced 2025-04-22; weights open-sourced Sept 2025 (pi05_base, pi05_libero, pi05_droid). Still the most capable open-weights PI model. Full fine-tuning needs >70 GB VRAM.

Sources: π0.5 blog, openpi, LeRobot docs.