Skip to content

Conversation

@jhen0409
Copy link
Collaborator

@jhen0409 jhen0409 commented Nov 5, 2023

This makes the server completion mode UI more like OpenAI Playground with complete mode - we can continue edit the result and use it as a new prompt in the same UI state. For me this is very convenient and fast for editing my prompts.

Demo:

Screen.Recording.2023-11-05.at.10.33.17.mov

Show probabilities are also works:
Screenshot 2023-11-05 at 10 35 25

I also disabled Markdownish in the completion mode, this make sure we can get the edit content correctly.

@jhen0409 jhen0409 force-pushed the server-completion-edit branch from bda68ea to 6c2c67d Compare November 6, 2023 00:09
@jhen0409 jhen0409 merged commit 4a4fd3e into ggml-org:master Nov 10, 2023
@jhen0409 jhen0409 deleted the server-completion-edit branch November 10, 2023 22:49
olexiyb pushed a commit to Sanctum-AI/llama.cpp that referenced this pull request Nov 23, 2023
* server : allow continue edit on completion mode

* server : handle abort case in runCompletion

* server : style improvement
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