Refact AI Self-Hosted

[Crack123451]

Hello. I have a problem deploying Refact AI on a server. The server does not have access to the Internet. I've found a command to run Self-hosted. But Refact AI is trying to get models from a remote storage. And I get an error. I can download the model manually to the server. But on the same server I do not have the opportunity to make edits to the git-project and create a docker image.

Question: does your project provide for launching a docker image specifying a folder with models (instead of a remote storage: https://huggingface.co/)?

[ser]

connect server to internet via a proxy

[olegklimov]

You can run it once on a computer that has internet access, verify the models you want work, and then take docker persistent volume to your new location. It will work just fine,

[BradKML]

Ollama for complete self-host, OpenRouter for options, LiteLLM in case there are anything missing... I wonder what else could make refact more flexible tho