refine nvfp4 algorithm #790
wenhuach21
merged
22 commits into
main
from
update_fp_torch_compile
update
45fd65db
Merge branch 'main' into update_fp_torch_compile
8821b470
![pre-commit-ci[bot]](https://avatars.githubusercontent.com/in/68672?v=4&s=60)
[pre-commit.ci] auto fixes from pre-commit.com hooks
4a2e7aa5
fix
6a489d26
Merge branch 'update_fp_torch_compile' of https://github.com/intel/au…
d82c018f
fix
37916fe0
recommend using iters 0 for nvfp
aa034023
force to enable `use_cache=False` in calibration
c82c933a
Merge branch 'main' into update_fp_torch_compile
6c8ef6b5
[pre-commit.ci] auto fixes from pre-commit.com hooks
8bf631db
fix pylint issue
f2e14729
rm torch compile in nv (exception for some models like phi4) and tune…
191fa900
Merge branch 'update_fp_torch_compile' of https://github.com/intel/au…
c58b3b52
update
436d1123
[pre-commit.ci] auto fixes from pre-commit.com hooks
c25d0326
update
c45746d2
Merge branch 'update_fp_torch_compile' of https://github.com/intel/au…
06ffa8a8
[pre-commit.ci] auto fixes from pre-commit.com hooks
da3a6f5d
fix rtn issue
e49a10bb
fix
a208adb4
Merge branch 'update_fp_torch_compile' of https://github.com/intel/au…
84ef3a2f
update
b6829263
wenhuach21
changed the title 
n1ck-guo
approved these changes
on 2025-09-05
Login to write a write a comment.
Login via GitHub