Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Update LLM server, fix bugs, and format with black #236

Merged
merged 10 commits into from
Nov 19, 2024
Merged

Conversation

jeremyfowers
Copy link
Collaborator

@jeremyfowers jeremyfowers commented Nov 18, 2024

This PR improves Turnkey-LLM in the following ways:

  1. Improvements to the LLM server:
    • Add /health, /stats, and /halt endpoints.
    • Add a constant, DEFAULT_SERVER_PORT to reference the server port without a magic number.
    • Catches the websocket disconnect error to allow clients to gracefully disconnect.
    • Hyperparameters now match the llm-prompt tool.
  2. Improve codebase consistency by requiring all PRs to be formatted with black

@jeremyfowers jeremyfowers self-assigned this Nov 18, 2024
@jeremyfowers jeremyfowers enabled auto-merge (squash) November 18, 2024 16:28
@gabeweisz
Copy link
Contributor

@ramkrishna2910 had asked me for a documentation update for the Llama.cpp PR - how about if I do that finish that PR before this one?

jeremyfowers and others added 3 commits November 18, 2024 14:10
Signed-off-by: Jeremy Fowers <80718789+jeremyfowers@users.noreply.github.com>
@jeremyfowers jeremyfowers changed the title Update LLM server, fix bugs, and add llamacpp support Update LLM server, fix bugs, and format with black Nov 19, 2024
@jeremyfowers jeremyfowers enabled auto-merge (squash) November 19, 2024 14:45
@jeremyfowers jeremyfowers merged commit 50996bb into main Nov 19, 2024
7 checks passed
@jeremyfowers jeremyfowers deleted the jfowers/server branch November 19, 2024 14:55
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.

3 participants