feature: dynamic quantized model support#1155
Draft
dsrenesanse wants to merge 1 commit intoml-explore:mainfrom
Draft
feature: dynamic quantized model support#1155dsrenesanse wants to merge 1 commit intoml-explore:mainfrom
dsrenesanse wants to merge 1 commit intoml-explore:mainfrom
Conversation
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Proposal: developers would like to use bigger and more intelligent models locally.
Problem: We need more flexibility for quantization to optimize memory footprint , but current state of mlx-lm allows to run a model only if all layers have same quantization.
Solution: be able to run models with aggressive quantization on knowledge layers for example, but with less aggressive quantization on reasoning and attention layers.
Edge cases: change of config file is required to support this feature.
Regards, Daniil.