Skip to content

Subfunction fixes for KV cache transform #655

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
vbaddi merged 5 commits into from
Dec 10, 2025
Merged

Subfunction fixes for KV cache transform #655

Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
d9f78d4
Subfunction fixes for KV cache transform
abhishek-singh591 Dec 8, 2025
dba4979
Resolved Lint and format issue
abhishek-singh591 Dec 8, 2025
b5d01a4
Resolved Lint and format issue
abhishek-singh591 Dec 8, 2025
c34522c
Resolved Lint and format issue
abhishek-singh591 Dec 8, 2025
cf09214
Merge branch 'quic:main' into subfunction_fix
abhishek-singh591 Dec 9, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
6 changes: 6 additions & 0 deletions QEfficient/base/onnx_transforms.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -19,16 +19,20 @@
from QEfficient.customop.ctx_scatter_gather import (
CtxGather,
CtxGather3D,
CtxGatherBlockedKV,
CtxGatherFunc,
CtxGatherFunc3D,
CtxGatherFuncBlockedKV,
CtxScatter,
CtxScatter3D,
CtxScatterFunc,
CtxScatterFunc3D,
)
from QEfficient.customop.ctx_scatter_gather_cb import (
CtxGatherBlockedKVCB,
CtxGatherCB,
CtxGatherCB3D,
CtxGatherFuncBlockedKVCB,
CtxGatherFuncCB,
CtxGatherFuncCB3D,
CtxScatterCB,
Expand Down Expand Up @@ -95,6 +99,8 @@ class CustomOpTransform(BaseOnnxTransform):
"CtxScatterFuncCB3D": (CtxScatterFuncCB3D, CtxScatterCB3D),
"CtxGatherFuncCB": (CtxGatherFuncCB, CtxGatherCB),
"CtxGatherFuncCB3D": (CtxGatherFuncCB3D, CtxGatherCB3D),
"CtxGatherFuncBlockedKV": (CtxGatherFuncBlockedKV, CtxGatherBlockedKV),
"CtxGatherFuncBlockedKVCB": (CtxGatherFuncBlockedKVCB, CtxGatherBlockedKVCB),
}

@classmethod
Expand Down
Loading