Download and run a local model
Go from choosing a local model to its first reply. Review and confirm the download, start llama-swap, then try a greeting for the fictional Aster Haven station.
Watch and learn
Video walkthroughs.
Each video follows one workflow from start to result, with English captions.
Browse the documentationCreated with pi-kit · English narration generated locally with Qwen3-TTS (Aiden). All example projects use fictional data.
9 walkthroughs
Explore an interactive view in chat
Ask for a fictional power planner, adjust its sliders, and reset the budget. A demo tool supplies the prebuilt HTML interface, rendered through pi-kit’s CopilotKit integration.
Choose an item with A2UI
Open a fictional supply checklist, inspect an item, and continue with its details. The tool-owned A2UI components send the selection back to the agent.
Inspect a web project
Open the fictional Orbit Board in Artifacts, inspect its Science filter, attach a screenshot to an unsent draft, and test the filter in the same browser.
Review file changes
Select code in a prepared fictional change, queue comments across two files, and send one feedback batch to pi. File references and selected code stay attached to the conversation.
Add a local provider
Connect an already-running local server, discover its model, save the provider, and send a test greeting to check the connection.
Find and add a skill
Find a public web-testing skill, review its instructions and source, install it, and add it to an unsent draft.
Install and use an extension
Review a pre-authored fictional package, install it, control whether it is enabled, and run its station-brief command. The fixed note comes from the extension, not a model reply.
Run background agents with Qwen 3.6
Delegate independent test planning and command-line review to two read-only background agents using Qwen 3.6 through LiteLLM. Inspect progress and bring their reports back into the main plan.