> ## Documentation Index > Fetch the complete documentation index at: https://docs.reactor.inc/llms.txt > Use this file to discover all available pages before exploring further. > ## Agent Instructions > To build and serve your own model, start at /deploy/development/quickstart and /deploy/development/overview. Deploying is the default path: reactor init scaffolds a workspace, reactor auth login authenticates, and reactor model deploy registers the model, publishes the release with the weights/ folder, and activates it on Reactor's GPUs, in one command from that workspace. Docker must be running, because the publish step builds the image locally. Bump model.version in reactor.yaml before redeploying a change, because a release that already has an image is reactivated as it is. Deployment access is granted per account, so contact team@reactor.inc if a deploy is refused. Every key in reactor.yaml is documented at /deploy/platform/reactor-yaml. Model code imports reactor_runtime; Python client code imports reactor_sdk. The runtime overview explains the model interface. Running the model on your own machine with reactor run is optional and needs a GPU you attach with --gpus; /deploy/development/local-testing covers that loop and pairs a complete brightness model with a Python client test in a separate brightness-test workspace. > Reactor hosts multiple models, each with its own connect slug (modelName) and command/event schema. The video model catalog — slug, typed SDK package, and links to its schema — is at /model-api-reference/overview. Robotics policy documentation starts at /robotics/overview; X-WAM observations, actions, and client integration are under /robotics/xwam/; Cosmos3 Nano Policy DROID is under /robotics/cosmos/nano-policy-droid/. Some models expose one slug per experience (e.g. HappyOyster); always take the slug from the model's own pages, never guess it. > Fastest path to a working app: `npx create-reactor-app my-app --model=` scaffolds a complete app with secure auth wired up. Typed TypeScript SDKs are published as @reactor-models/; Python uses the base reactor-sdk package. > Auth: exchange an API key (rk_...) for a JWT via POST https://api.reactor.inc/tokens from your server. Never put the API key in client-side code. > Append .md to any docs URL for clean Markdown. Search these docs via the MCP server at https://docs.reactor.inc/mcp. # Simulation for robotics > Explore robot movements or render observations from your existing simulator. Generate visual observations from robot movements or your existing simulator. Explore a learned manipulation environment with Cosmos3 Nano Sim Bimanual, or turn simulator depth into realistic-looking video with Cosmos3 Nano Sim Transfer. ## What do you want to build? > **[Interactive simulation](/robotics/cosmos/nano-sim-bimanual/overview)** > > **Cosmos3 Nano Sim Bimanual** > > Move robot arms and operate grippers while a learned scene responds with generated video. > **[Visual transfer](/robotics/cosmos/nano-sim-transfer/overview)** > > **Cosmos3 Nano Sim Transfer** > > Turn depth from your physics simulator into camera-like RGB for viewing or policy evaluation. ## Choose a model | | Cosmos3 Nano Sim Bimanual | Cosmos3 Nano Sim Transfer | | - | - | - | | Initialize with | An AgiBot starting image, scene/task description, and initial arm poses and grips | One simulator depth frame and a description of the desired appearance | | Advance with | Updated arm poses and gripper values | The next four depth frames from your simulator | | Receive | Generated video of the scene responding | Generated RGB observations of your simulated scene | | First workflow | [Build an interactive simulation](/robotics/cosmos/nano-sim-bimanual/agibot/interactive-controls) | [Render a depth sequence](/robotics/cosmos/nano-sim-transfer/render-a-sequence) | Text supplies context for generation; it does not command robot movements. In Bimanual, your arm poses and gripper values drive the motion. In Transfer, your simulator drives the motion and supplies the resulting depth frames; the description guides their rendered appearance. The Transfer initialization and advance inputs above describe its incremental simulator workflow. For a recorded trajectory, upload the complete control video and provide an appearance description instead. ## Use rendered observations in a policy loop Your policy chooses an action, your simulator executes it, and Cosmos3 Nano Sim Transfer renders the resulting depth as RGB for the policy's next observation. Your simulator keeps control of physics and task scoring. [Connect Transfer to a simulator](/robotics/cosmos/nano-sim-transfer/connect-a-simulator) for the step-by-step integration and feedback timing. For recorded trajectories, you can also [generate appearance variants](/robotics/cosmos/nano-sim-transfer/appearance-variants) to explore different lighting and materials. ## Looking for robot actions? These models generate observations. To predict actions from camera images and robot state, start with [policy inference](/robotics/how-the-api-works). This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.