π0 (pi-zero)
Announced 2024-10-31; weights released 2025-02-04 (openpi). Not the first open VLA (OpenVLA/Octo came earlier) but became the most widely used open generalist robot policy baseline. Superseded by π0.5; still available. Fine-tuned expert checkpoints: pi0_droid, pi0_aloha_towel, pi0_aloha_tupperware, pi0_aloha_pen_uncap.
- Input
- text, image
- Output
- action
- License
- apache-2.0 (openpi code); weights built on PaliGemma, Gemma terms apply
- Verified
- 2026-09-29
How to call it
| Provider | Model id | Endpoint / URL | Docs |
|---|---|---|---|
| GitHub (openpi, JAX + PyTorch) | gs://openpi-assets/checkpoints/pi0_base | github.com/Physical-Intelligence/openpi | — |
| Hugging Face (LeRobot port) | lerobot/pi0_base | huggingface.co/lerobot/pi0_base | docs |
Notable capabilities (2)
- Flow-matching VLA for dexterous, high-frequency control: PaliGemma VLM plus an action expert that outputs continuous action chunks via flow matching (up to 50 Hz), trained on data from 8 distinct robots; folds laundry, busses tables, assembles boxes. source
- Open weights with fine-tuning recipes (found after launch): Open-sourced 2025-02-04 in openpi with base and fine-tuned checkpoints (ALOHA towel/tupperware/pen, DROID) pre-trained on 10k+ hours of robot data; inference needs >8 GB VRAM, LoRA fine-tuning >22.5 GB. source
Physical Intelligence's first generalist robot policy and the base of the open-source openpi stack.
Sources: π0 blog, Open Sourcing π0, openpi, The Robot Report.
Other Physical Intelligence models
π0.7 · π0.5 · π0.6 / π*0.6 · π0-FAST