Merged
Conversation
juliendenize
force-pushed
the
improve_llama_cpp_integration
branch
3 times, most recently
from
91435d4 to
e85ebee
Compare
juliendenize
force-pushed
the
improve_llama_cpp_integration
branch
from
e85ebee to
ec07ed7
Compare
patrickvonplaten
approved these changes
Jul 25, 2025
Contributor
patrickvonplaten
left a comment
patrickvonplaten
left a comment
There was a problem hiding this comment.
Some nits for docs
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
This PR adds a FastAPI app server to
mistral-common.The features include:
apply a "chat template". We actually don't have chat templates in mistral-common but a dedicated route will return the same string as a chat template would.Not gonna do that as discussed on Slack because might be prone to issues (we should tokenize always as ints)This should improve integrations with some LLM inference backends such as llama.cpp.
Edit:
Now also supports tool call parsing via the detokenize route when passing to the request
as_message=True