common: fix sub-model HF download when spec-draft-hf is specified - #25039
Closed
ggerganov wants to merge 1 commit into
Closed
common: fix sub-model HF download when spec-draft-hf is specified#25039ggerganov wants to merge 1 commit into
ggerganov wants to merge 1 commit into
Conversation
The new common_models_handler only built an HF plan for the main model, ignoring sub-model HF repos like --spec-draft-hf. Ref: #24980 (comment) Assisted-by: pi:llama.cpp/Qwen3.6-27B
Member
Author
|
@ngxson I had a bit of trouble updating the logic so I let the Qwen come up with something. This seems to work, but if there is a better way to do it - feel free to ignore this PR. In general, I feel like we should somehow rework the model downloads further - there are many different use cases and it is getting difficult to keep track of all the possible configs. Btw, we will also want to add eagle3 and dflash sidecars too, similar to mtp. |
Collaborator
|
Thanks, I completely forgot about this case (as well as |
Collaborator
|
Can you try #25043 ? |
Member
Author
|
Superseded by #25043 |
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Overview
cont #24980
fix #24980 (comment)
Fix sub-model HF downloads (draft, mmproj, vocoder) when they have their own
hf_repospecified. Thecommon_models_handlerrefactor only built an HF plan for the main model, ignoring--spec-draft-hfand similar options.Additional information
Before this fix, configs like:
Would fail because the draft model's HF repo was never downloaded — the path remained as a bare filename instead of the full cache path.
Requirements