The AI services are currently offline while I work on the backend, hardware, and models.
I've started running some much larger models, including 20B+ models, and the current hardware wasn't really designed to handle all of them.
Rather than keep the service online while I'm moving things around, I'm taking it offline for a while so I can work on everything without people accidentally using it during development.
I'm currently working on the models, infrastructure, performance, and hardware setup. The AI endpoint is currently limited to local testing.
The AI services are planned to return on:
The goal is to eventually make the AI services publicly available once everything is ready.
Canyon Lake Labs
AI Services