Model card
- Made by
- Physical Intelligence
- Size
- 3B parameters
- Published
- 22 Apr 2025
More about the model
- Publication
- arXiv 2504.16054
- Weights licence
- Apache 2.0, with the Gemma Terms of Use source
- Smallest GPU
- More than 8 GB of GPU memory, for example an RTX 4090 source
- Inference time
- About 76 ms per chunk on an RTX 4090 (bfloat16, 5 denoising steps) source
Run with Positronic
π₀.₅ DROID
Start the server source
docker run --gpus all -p 8000:8000 positro/pi05-droidDrive a DROID robot from a positronic checkout guide
uv run --extra hardware positronic-inference real \
--policy=.remote \
--policy.address.host=localhost --policy.address.port=8000 \
--next_task.instruction="Put the marker on the plate"Progress through a run
Each band shows how many runs of π₀.₅ have placed that number of items at each second. The dashed line shows the mean share of placed items per run.
Loading runs...
Episodes
Each episode of π₀.₅ on Pick and Place, the newest first.
Progress through an episode
Each band shows how many episodes of π₀.₅ had succeeded, partly succeeded (some subtask score) or had no success at each second of simulated time. An episode that has ended stays in its band. The dashed line shows the mean Reliability, the mean subtask score.
Episodes
Each episode of π₀.₅ on RoboLab-120, the newest first.
Progress through an episode
Each band shows how many episodes of π₀.₅ had succeeded or had no success at each second of simulated time. An episode that has ended stays in its band. The dashed line shows the mean Reliability, the share of the episodes that had succeeded.
Episodes
Each episode of π₀.₅ on MolmoSpaces-320, the newest first.