physical-ai-studio
Physical AI Studio is an end-to-end framework for training robots to perform tasks through imitation learning from human demonstrations.
Safety scan
A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.
Check it deeper on your machine
It has changed since its scan on 11 Nov 2026.
As a Claude Code plugin, install it and then ask Claude Code:
/plugin marketplace add keithmackay/mackayi/plugin install scanskillsafety@mackayiDo a deep scan of https://github.com/open-edge-platform/physical-ai-studio with scanskillsafety
Or from a terminal, clone it and run it:
git clone https://github.com/keithmackay/scanskillsafety-skill ~/.claude/skills/scanskillsafetynode ~/.claude/skills/scanskillsafety/cli.cjs --deep 'https://github.com/open-edge-platform/physical-ai-studio'
It checks the exact commit: every file, its packages against OSV (api.osv.dev), how it compares with this page's scan (findsafeskills.com receives the repository's address), and, if you agree, Claude reading every file with your own Claude subscription. It never runs the code.
Badge for your README
Markdown
[](/listings/open-edge-platform--physical-ai-studio)HTML
<a href="/listings/open-edge-platform--physical-ai-studio"><img src="/api/listings/open-edge-platform--physical-ai-studio/badge.svg" alt="findsafeskills safety rating" /></a>Install
Install instructions from the README
As written by the repo's author. findsafeskills has not verified these steps.
Application (GUI)
For users who prefer a visual interface for end-to-end workflow:
<p align="center"> <a href="docs/assets/physical_ai_studio_full_overview.mp4"> </a> </p>
[Download the full demo video →](docs/assets/physical_ai_studio_full_overview.mp4)
[Application Documentation →](./application/README.md)
Docker
Run the full application (backend + UI) in a single container (using [Docker](https://docs.docker.com/engine/install/ubuntu/)):
# Clone the repository git clone https://github.com/open-edge-platform/physical-ai-studio.git cd physical-ai-studio # Setup and run docker services cd application/docker ./setup-devices.sh --xpu # or use --cuda, --cpu docker compose up -d
Application runs at <http://localhost:7860>. See the [Docker README](./application/docker/README.md) for hardware configuration (Intel XPU, NVIDIA CUDA) and device setup.
If you plan to train Hugging Face Hub-backed policies (for example, SmolVLA, Pi0.5, and others), configure `HF_TOKEN` to avoid unauthenticated Hub access warnings. See [Hugging Face Integration](./application/backend/docs/huggingface_integration.md).
Native: installation & running
Run the application in development mode, using [uv package manager](https://docs.astral.sh/uv/getting-started/installation/) and [node v24](https://nodejs.org/en/download) (we recommend using nvm)
Note: native setup requires additional OS-level libraries (OpenCV/video/USB and Python build dependencies). See the **Prerequisites** section in [Application Installation](./application/docs/01-installation.md#prerequisites).
# Clone the repository git clone https://github.com/open-edge-platform/physical-ai-studio.git cd physical-ai-studio # Install and run backend cd application/backend # Start the backend, or use --extra cpu, --extra cuda uv run --extra xpu physicalai-studio serve # or: ./run.sh
# In a new terminal: install and run UI cd application/ui npm install # Start the UI npm run start
Open <http://localhost:3000> in your browser.
If you plan to train Hugging Face Hub-backed policies (for example, SmolVLA, Pi0.5, and others), configure a Hugging Face token in Settings > General > Hugging Face in the UI to avoid unauthenticated Hub access warnings. See [Hugging Face Integration](./application/backend/docs/huggingface_integration.md).
Library (Python/CLI)
For programmatic control over training, benchmarking, and deployment with both API and CLI
pip install physicalai-train
<details open> <summary>Training</summary>
from physicalai.data import LeRobotDataModule from physicalai.policies import ACT from physicalai.train import Trainer datamodule = LeRobotDataModule(repo_id="lerobot/aloha_sim_transfer_cube_human") model = ACT() trainer = Trainer(max_epochs=100) trainer.fit(model=model, datamodule=datamodule)
</details>
<details> <summary>Benchmark</summary>
from physicalai.benchmark import LiberoBenchmark
from physicalai.policies import ACT
policy = ACT.load_from_checkpoint("experiments/lightning_logs/version_0/checkpoints/last.ckpt")
benchmark = LiberoBenchmark(task_suite="libero_10", num_episodes=20)
results = benchmark.evaluate(policy)
print(f"Success rate: {results.aggregate_success_rate:.1f}%")</details>
<details> <summary>Export</summary>
from physicalai.export import get_available_backends
from physicalai.policies import ACT
# See available backends
print(get_available_backends()) # ['onnx', 'openvino', 'torch', 'executorch']
# Export to OpenVINO
policy = ACT.load_from_checkpoint("experiments/lightning_logs/version_0/checkpoints/last.ckpt")
policy.export("./policy", backend="openvino")</details>
<details> <summary>Inference</summary>
What was scanned
The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.
README.md · README · 170 lines
View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)
This file has changed since it was scanned, so this is not exactly the text that was scanned, and no line is marked. The findings are listed with the safety scan above.
- <p align="center">
- <img src="docs/assets/physical_ai_studio.png" alt="Physical AI Studio" width="100%">
- </p>
- <div align="center">
- **Physical AI Studio helps teams train, benchmark, and deploy robot policies from demonstrations.**
- [Studio UI](#studio-ui) • [Python API](#python-api) • [CLI](#cli) • [Policies](#policies) • [Install](#install) • [Documentation](#documentation)
- </div>
- ---
- ## Studio UI
- Collect demonstrations, train policies, and run inference through a visual interface.
- <!-- markdownlint-disable MD033 -->
- <p align="center">
- <img src="docs/assets/readme/gui-demo.gif" alt="Physical AI Studio UI demo: configure robots and cameras, record data, train, review models, and deploy to robots" width="100%">
- </p>
- <!-- markdownlint-enable MD033 -->
- [Start Studio with Docker](./application/docs/01-installation.md#install-with-docker-recommended) · [Native setup](./application/docs/01-installation.md#alternatively-install-natively)
- <details>
- <summary>Copy Docker start commands</summary>
- ```bash
- git clone https://github.com/open-edge-platform/physical-ai-studio.git
- cd physical-ai-studio/application/docker
- cp .env.example .env
- ./setup-devices.sh --cpu # or --xpu or --cuda
- docker compose up -d
- ```
- Open <http://localhost:7860>. See the [Docker guide](./application/docker/README.md) for hardware options and setup. Configure `HF_TOKEN` in `application/docker/.env` for Hugging Face-backed policies; see [Hugging Face integration](./application/backend/docs/huggingface_integration.md).
- </details>
- <details>
- <summary>Native development commands</summary>
- Native setup is for contributors and requires OS-level libraries, [uv](https://docs.astral.sh/uv/getting-started/installation/), and [Node.js v24](https://nodejs.org/en/download). See the [prerequisites](./application/docs/01-installation.md#prerequisites).
- In one terminal, from the repository root:
- ```bash
- cd application/backend
- uv run --extra xpu physicalai-studio serve # or --extra cpu / --extra cuda
- ```
- In a second terminal, also from the repository root:
- ```bash
- cd application/ui
- npm install
- npm run start
- ```
- Open <http://localhost:3000>. For native Hugging Face setup, use **Settings > General > Hugging Face**.
- </details>
- ## Python API
- Train, benchmark, and export policies programmatically.
- <!-- markdownlint-disable MD033 -->
- <p align="center">
- <img src="docs/assets/readme/api-demo.gif" alt="Python API demo: train, benchmark, export, and deploy a policy" width="100%">
- </p>
- <!-- markdownlint-enable MD033 -->
- The library also supports LIBERO and RoboCasa benchmarks; see the [benchmarking guide](./library/docs/getting-started/first-benchmark.md) for suite-specific setup and checkpoint requirements.
- <details>
- <summary>Copy the Python example</summary>
- ```python test="skip" reason="requires dataset download and robot observation"
- from physicalai.benchmark.gyms import PushTBenchmark
- from physicalai.data import LeRobotDataModule
- from physicalai.inference import InferenceModel
- from physicalai.policies import ACT
- from physicalai.train import Trainer
- datamodule = LeRobotDataModule(repo_id="lerobot/pusht", num_workers=0)
- policy = ACT()
- trainer = Trainer(max_epochs=100)
- trainer.fit(model=policy, datamodule=datamodule)
- benchmark = PushTBenchmark(num_episodes=50)
- results = benchmark.evaluate(policy)
- policy.export("./exports/act", backend="openvino")
- runtime_policy = InferenceModel("./exports/act")
- ```
- </details>
- A PushT-trained policy is not directly usable on a physical robot; use a policy trained for that robot with a matching Runtime config. See the [robot deployment guide](https://github.com/openvinotoolkit/physicalai/blob/main/docs/getting-started/run-a-policy.md).
- ## CLI
- Run Studio training, benchmarking, and export from Runtime's `physicalai` CLI; Runtime also provides the `run` command for deployment.
- <details>
- <summary>Copy the CLI commands</summary>
- The example configs live in `library/configs`; run these Studio commands from `library/` in a cloned repository:
- ```bash
- # From the repository root
- cd library
- # Train on PushT
- physicalai fit \
- --config configs/physicalai/act/pusht/default.yaml
- # Benchmark the trained checkpoint
- physicalai benchmark \
- --config configs/benchmark/pusht.yaml \
- --policy physicalai.policies.ACT \
- --ckpt_path experiments/lightning_logs/version_0/checkpoints/last.ckpt
- # Export for deployment
- physicalai export \
- --policy physicalai.policies.ACT \
- --ckpt_path experiments/lightning_logs/version_0/checkpoints/last.ckpt \
- --backend openvino \
- --output_dir exports/act
- ```
- From your Runtime project, run a robot policy with:
- ```bash
- physicalai run --config runtime.yaml
- ```
- Use the [Runtime config example](https://github.com/openvinotoolkit/physicalai/blob/main/docs/how-to/config/write-runtime-config.md) as `runtime.yaml`; it is separate from the PushT training config above.
- </details>
- ## Policies
- Native policy families include ACT, Pi0.5, SmolVLA, MolmoAct2, RLDX-1, XR0, and Cosmos3; see the library’s [supported policies](./library/README.md#supported-policies).
- ## Install
- Use the library for Python API and CLI workflows, or follow the Studio UI setup above. The library supports Python 3.12–3.14.
- See the [library installation guide](./library/docs/getting-started/installation.md) for source installs and optional dependencies.
- <details>
- <summary>Copy the Python install command</summary>
- ```bash
- pip install physicalai-train
- ```
- For hardware-specific PyTorch packages, use the source installation instructions
- in the [library installation guide](./library/docs/getting-started/installation.md).
- </details>
- ## Documentation
- [Library](./library/README.md) · [Application](./application/README.md) · [Contributing](./CONTRIBUTING.md)
Ratings and comments
Sign in to rate this skill.
Sign inFree to use. If it’s helped, pass it on.