Extra_options disable_qkv_fusion to untie qkv_projs from upstream choice
#1893
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
kunal-vaishnavi
previously approved these changes
Nov 26, 2025
jixiongdeng
force-pushed
the
jd/disable_fuse_qkv
branch
from
e4ebe46 to
9658ae7
Compare
Contributor
Author
|
Rebased main & resolved conflict. Thank you! @kunal-vaishnavi |
tianleiwu
approved these changes
Nov 26, 2025
kunal-vaishnavi
pushed a commit
that referenced
this pull request
Dec 5, 2025
…hoice (#1893) ## Problem As we discussed in [this PR](#1885), I separate `disable_qkv_fusion` option as a new PR. The current model builder ties q_proj, k_proj and v_proj together as qkv_proj by default, which is not controllable by upstream quantization choice. ## Solution Added `disable_qkv_fusion` in extra_options to override `attention_attrs["use_packed_matmul"]`. Running examples: **untied qvk_projs for 4 bit rtn on Llama-3.2-3B-Instruct**: ``` python src/python/py/models/builder.py -m meta-llama/Llama-3.2-3B-Instruct -p int4 -e cuda -o export_model/llama32_3bi_rtn_u4_untied_qkv --extra_options int4_algo_config=rtn disable_qkv_fusion=true ``` ## Changes ### Modified Files - `src/python/py/models/builder.py` - `src/python/py/models/builders/base.py` - `src/python/py/models/README.MD` ### Key Modifications 1. Added `disable_qkv_fusion` as a part of assigning logic of `attention_attrs["use_packed_matmul"]`. 2. Added documents.
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Outdated suggestions cannot be applied.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.
Problem
As we discussed in this PR, I separate
disable_qkv_fusionoption as a new PR.The current model builder ties q_proj, k_proj and v_proj together as qkv_proj by default, which is not controllable by upstream quantization choice.
Solution
Added
disable_qkv_fusionin extra_options to overrideattention_attrs["use_packed_matmul"].Running examples:
untied qvk_projs for 4 bit rtn on Llama-3.2-3B-Instruct:
Changes
Modified Files
src/python/py/models/builder.pysrc/python/py/models/builders/base.pysrc/python/py/models/README.MDKey Modifications
disable_qkv_fusionas a part of assigning logic ofattention_attrs["use_packed_matmul"].