Mozilla-Ocho / llamafile

Distribute and run LLMs with a single file.
https://llamafile.ai
Other
20.58k stars 1.04k forks source link

Support for configurable URL prefix when running in server mode #597

Closed vlasky closed 1 month ago

vlasky commented 1 month ago

Added support for new commandline parameter --url-prefix that sets the URL prefix (subdirectory) under which the API will be served when llamafile is running in server mode, e.g.

llamafile --server --url-prefix /llamafile -m mymodel.gguf

Makes it much easier to run llamafile behind a reverse proxy.