Skip to content

Conversation

@mostlygeek
Copy link
Owner

@mostlygeek mostlygeek commented Dec 17, 2024

This adds an /upstream/:model_name endpoint which is a reverse proxy for the loaded inference server. Useful for accessing the UI of the upstream, or other endpoints, if it has one.

New functionality:

  • automatically load/swap the model depending on the ID in the URL
  • /upstream shows an index of available models

llama-swap-upstream-demo

This PR also introduces the unlisted: true configuration for models to omit them from /v1/models and /upstream.

- add /upstream endpoint to show a list of available models
- add `unlisted` configuration option to omit a model from /v1/models
  and /upstream lists
@mostlygeek mostlygeek merged commit 891f6a5 into main Dec 17, 2024
@mostlygeek mostlygeek deleted the add-upstream branch December 17, 2024 22:37
@mostlygeek mostlygeek changed the title Add /upstream endpoint (#30) Add /upstream endpoint Dec 17, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants