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

Example UI to "play" with the API #6

Closed
vidas opened this issue Dec 11, 2024 · 1 comment · Fixed by #12
Closed

Example UI to "play" with the API #6

vidas opened this issue Dec 11, 2024 · 1 comment · Fixed by #12
Assignees
Labels

Comments

@vidas
Copy link
Member

vidas commented Dec 11, 2024

It would be great to have a UI with which users could test out the API server (even though it will be mostly accessed through the code). The main goal is to have something up and running in seconds. The barrier to entry should be as low as possible. Something as low effort as running make demo.

(Other uses - actual demos and testing/debugging stuff by developers).

  1. Please research open source LLM Web UI tools (like Chat UI by HuggingFace or Open WebUI.
  • selected UI must use OpenAI API and have configurable URL
  • selected UI must use /v1/chat/completions API endpoint (most use legacy and low-level /v1/completions)
  • it should be possible (preferably via user interface) to tweak various parameters, like max_tokens, temperature, top_n and so on.
  1. Please share your findings.
  2. Create a script in the examples folder/makefile target to run the preconfigured UI (container) alongside our example NekkoAPI server container.
@vidas vidas added the chore label Dec 11, 2024
@akxcv akxcv self-assigned this Dec 12, 2024
@akxcv
Copy link
Collaborator

akxcv commented Dec 12, 2024

Open WebUI satisfies all the listed requirements and is quite easy to set up. Posted a demo implementation in #12.

One downside I found is poor control over playground features: there's no way to disable speech recognition or other capabilities that our server does not yet support. Though I don't think that's a big deal, since our goal is to implement full compatibility with OpenAI API. Also, it won't be difficult to switch our UI to something else down the line if needed.

@akxcv akxcv closed this as completed in #12 Dec 13, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Projects
None yet
Development

Successfully merging a pull request may close this issue.

2 participants