add support block-wise quant from bf16 #925
Open
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
This was referenced Jul 1, 2025
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
add native bf16 to fp8 block-wise quant utility for NV E4M3 data type
Currently many toolkits do not support block-wise quantization, (see the recently raised llmcompressor issue PR#1475) and torch ao topics.
Since former release of DeepV3 (03-24), more and more entities are working on alignment of DeepSeekV3, and bf16 training is very handy in major training framework.
It is native to add this processing script here so that people can immediately quant the model from bf16.
Procedure how to use it
please see inference/README.md
Correctness Verification (H800 2x8 setting up)
setup model:
send a query