common,server : fix custom preset dedup against cached models#25235
Open
angt wants to merge 1 commit into
Open
common,server : fix custom preset dedup against cached models#25235angt wants to merge 1 commit into
angt wants to merge 1 commit into
Conversation
Revert the canonical_tag() case-folding in preset.cpp that collapsed distinct alias presets (e.g. Q4_K_XL-thinking and Q8_0-thinking into THINKING) and broke downstream tools keying by name. Instead, add cached_base_name() in server-models.cpp so a custom preset section whose quant tag carries a convenience prefix merges into the cached entry the cache stores without that prefix. Presets whose tag suffix is not a cached quant keep their own name as a separate alias row. Also track custom_names during the merge so the '*' marker lands on the actual mapping key rather than the unmerged custom name.
Collaborator
|
IMO it might be less error-prone if the dedup logic firstly resolves the model to exact local path. I'll will try this direction and push a fix if it works |
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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
Revert the canonical_tag() case-folding in preset.cpp that collapsed distinct alias presets (e.g. Q4_K_XL-thinking and Q8_0-thinking into THINKING) and broke downstream tools keying by name.
Instead, add cached_base_name() in server-models.cpp so a custom preset section whose quant tag carries a convenience prefix merges into the cached entry the cache stores without that prefix. Presets whose tag suffix is not a cached quant keep their own name as a separate alias row.
Also track custom_names during the merge so the '*' marker lands on the actual mapping key rather than the unmerged custom name.
Additional information
This is another solution that is #25226.
Requirements