Bf16 support for tinychat 2.0 on GEMV and GEMM #255
Conversation
This file contains 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
|
Hi @jason-huang03 . Thank you for the great work! I will approve all the changes till now. Do we have any modification for the front-ends in accordance with the bf16 support. I guess it would be better to merge them to the codebase within the same PR. 😊 |
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Outdated suggestions cannot be applied.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.
This PR adds bf16 support for GEMV and GEMM operation under
awq/kernels/csrc/quantization_newdirectory. BF16 is crucial for serving some model like Qwen2.5 72B, which would encounter NaN problem using fp16 even with original model.The main point of this PR is: