VLAOpen

pi0 / OpenPI

Physical Intelligence

Generalist robot policy family and open-source robotics model package from Physical Intelligence.

Code: Available
Weights: Available
Checkpoint: Available
License: apache-2.0

Weights Code Model index A–Z

Decision summary

Best for

Generalist policy transfer

Main blocker

Hardware, latency, dependencies, and checkpoint loading are not pipeline-tested.

Next check

Run a checkpoint loading smoke test before marking the model pipeline-tested.

Release history

Source-backed release timing for this canonical model record.

Important releases

2024-10-21
Model series introduced
Release timing is recorded from the official source creation date.
OpenPI

Evidence profile

A visual read of adoption evidence. Scores describe catalog evidence readiness, not task performance.

Published evaluation

5/6

dimensions scored

Unknown evidence remains visible and is never treated as a zero.

Access and governance75conf. 85

Artifact availability90conf. 90

Training and loading reproducibility70conf. 65

Training data requirements76conf. 70

Evaluation evidenceNot scoredconf. 10

Deployment readiness70conf. 55

Why these scores

The strongest decision reasons behind the evidence profile.

Top 3 adoption signals

Artifact availability

Useful

Checkpoint files were verified in the official repository metadata.

90 · confidence 90

Training and loading reproducibility

Useful

Configuration files are present; runtime loading is not tested.

70 · confidence 65

Evaluation evidence

Unknown

No structured evaluation evidence has been verified.

— · confidence 10

Data requirements

Declared loop-data needs, missing evidence, and the next checks that matter.

4 required categories

Observation / ego video

observation

required

Language intent / task phase

language intent

required

Action / robot state

actions · robot state

required

Sim-real / embodiment metadata

robot state · embodiment

required

Critical gaps

Core categories are represented. Interface alignment and data quality still require verification.

Linked datasets

Signal-level links only. Verify runtime interfaces before use.

Browse datasets

EgoWorld

Schema validation for LeRobot v3

Useful

provisional fit 83 · confidence 40

2 matched · 1 unknown

Manus Egocentric Sample

Testing LeRobot import paths

Useful

provisional fit 83 · confidence 40

2 matched · 1 unknown

EgoStation GoPro Pick-and-Place

Pick-and-place benchmark fixtures

Useful

provisional fit 100 · confidence 60

3 matched · 0 unknown

More evidenceShow

Additional decision dimensions

Access and governanceUseful

Access and license are declared by the source.

75 · c85

Training data requirementsUseful

Required signal categories are structured; exact tensor and action interfaces still need verification.

76 · c70

Deployment readinessUseful

Hardware, latency, dependencies, and checkpoint loading are not pipeline-tested.

70 · c55

Artifact facts

release_year: official_claim; 2024; curated official source year
model.required_signals: official_claim; observation · language intent · actions · robot state · embodiment; Official model documentation and curated signal mapping
repository.code: metadata_verified; repository: Physical-Intelligence/openpi · default branch: main · archived: false · code file count: 97 · sample files: examples/aloha_real/constants.py · examples/aloha_real/convert_aloha_data_to_lerobot.py · examples/aloha_real/env.py · examples/aloha_real/main.py · examples/aloha_real/real_env.py · examples/aloha_real/requirements.txt · examples/aloha_real/robot_utils.py · examples/aloha_real/video_display.py · examples/aloha_sim/env.py · examples/aloha_sim/main.py · examples/aloha_sim/requirements.txt · examples/aloha_sim/saver.py · examples/convert_jax_model_to_pytorch.py · examples/droid/compute_droid_nonidle_ranges.py · examples/droid/convert_droid_data_to_lerobot.py · examples/droid/main.py · examples/libero/convert_libero_data_to_lerobot.py · examples/libero/main.py · examples/libero/requirements.txt · examples/simple_client/main.py; GitHub repository metadata and recursive tree
model.hub: official_claim; declared; curated model hub link
model.weight_files: metadata_verified; repository: lerobot/pi0 · count: 1 · sample files: model.safetensors; Hugging Face model metadata siblings
model.checkpoint_files: metadata_verified; repository: lerobot/pi0 · count: 1 · sample files: model.safetensors; Hugging Face model metadata checkpoint files
model.config_files: metadata_verified; repository: lerobot/pi0 · count: 1 · sample files: config.json · repository: physical-intelligence/fast · count: 2 · sample files: processor_config.json · tokenizer_config.json; Hugging Face model metadata siblings
model.training_data: official_claim; robot data hours: 10k+ · applies to: pi0 · pi0-FAST · pi0.5 base checkpoints; README, base model checkpoints
model.runtime_requirements: official_claim; inference gpu memory: >8 GB · lora finetuning gpu memory: >22.5 GB · full finetuning gpu memory: >70 GB · tested os: Ubuntu 22.04; README, requirements
license.weights: metadata_verified; apache-2.0; Hugging Face model card license metadata
license.code: metadata_verified; Apache-2.0; GitHub repository SPDX license metadata
license: metadata_verified; apache-2.0; Hugging Face model card license metadata
release_timing: metadata_verified; 2024-10-21T15:23:28.000Z; GitHub repository metadata created_at

Additional references

pi0 postPost FAST tokenizerModel hub

OpenBot notes

Useful for framing OpenBot as a data-readiness layer around emerging generalist policies.
Dataset pages should expose whether a source can become a fine-tuning trace, not just whether it has video.