Skip to main content

Install the reference client

New to Reactor? How the API works explains the session, camera streams, commands, and action messages used below. You need Python 3.12, uv, a Reactor API key, and Reactor’s public cookbook. The client code is public; your Reactor account still needs access to the selected hosted model. No local model weights or GPU are required for this example.
Run the example below from robotics/sim/notebooks/. This environment installs the Python SDK and helpers from the Robotics Cookbook for camera publishing and message queues. The cookbook helpers are separate from the SDK. Keep your API key in the environment. On Linux, the published glibc wheel requires glibc 2.34 or newer. The default endpoint is https://api.reactor.inc; REACTOR_API_URL overrides it. Examples select a model by slug, not a frozen release; check its release notes when the deployed contract changes. If setup or connection fails, open your model guide and follow its troubleshooting page.

Request a chunk

Run the cookbook example from the current robotics/sim/notebooks/ directory:
Expect actions: (24, 8) and eight numeric values for the first row. The synthetic observations check connectivity and reply shape; do not execute their outputs on hardware. The helper repeatedly publishes the test images for this check. When connecting cameras, send new images as you capture them: the model uses recent frames to understand motion. The 15 fps setting controls how often images are published; your robot controller sets the action execution rate. See integration. The printed prompt-to-chunk seconds measures the time from sending the task prompt to receiving the first chunk. The timer starts after session setup and the initial half-second camera-frame wait. inference_seconds measures model inference alone, excluding network transfer. Waiting for frames does not guarantee that camera images and robot state were captured together. Next, run RoboLab or use the integration guide. For missing replies, see troubleshooting.