simplexai-labs/PolyStrata ? reverse-engineered prompt
Reverse engineered prompt
Build me a desktop app and local server that lets me run big AI models on my own PC with no cloud calls. I want a simple one click installer for Windows and Linux, and after setup I should be able to launch it, pick a supported model, and chat with it from a browser or terminal. It should expose a localhost API that works like OpenAI and Anthropic so I can point coding tools at it. Please include a clean way to switch between installed models while the server is running, a basic monitor that shows what the model is doing, and an easy setup flow for Nvidia and AMD cards where supported. Make sure it can handle text and image input for the models that support that, and include a small benchmark or status tool so I can see speed and memory use. If you need to check current docs for model or driver details, look them up online first.
Are you gonna build this?
make sure you review the code using coderabbit