swap model to Qwen2.5 0.5B for CPU-friendly demo #2
Loading…
Add table
Add a link
Reference in a new issue
No description provided.
Delete branch "qwen2.5-model-swap"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
This PR swaps the demo from Phi-3.1-mini to a smaller Qwen2.5 0.5B model. The goal is a lower-risk, CPU-friendly demo rather than maximizing model quality.
It also updates local dev behavior so the model download is checked, browser testing works against the local llama.cpp server, and the chat UI streams responses so users see text appear instead of waiting on the typing dots.
While I was at it, I also refreshed the visible model copy in the chat window footer, and updated the Caution CLI link so it links to the correct CLI repo.
View command line instructions
Checkout
From your project repository, check out a new branch and test the changes.Merge
Merge the changes and update on Forgejo.Warning: The "Autodetect manual merge" setting is not enabled for this repository, you will have to mark this pull request as manually merged afterwards.