Skip to content

Avoid squeezing original weight tensors with leading dim 1 #294

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
merged 4 commits into from
Sep 9, 2025
Merged

Avoid squeezing original weight tensors with leading dim 1 #294

Changes from 2 commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
867922a
Do not squeeze weights with leading dim 1
cjluo-nv Sep 5, 2025
76ce70a
Update quant_utils.py
cjluo-nv Sep 5, 2025
501d8f0
Update quant_utils.py
cjluo-nv Sep 5, 2025
ab4e6d9
Merge branch 'main' into cjluo-nv-patch-1
cjluo-nv Sep 8, 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
11 changes: 8 additions & 3 deletions modelopt/torch/export/quant_utils.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -869,11 +869,16 @@ def postprocess_state_dict(state_dict: dict, maxbound: float, quantization: str
post_state_dict[prefix + new_suffix] = value
break

# Squeeze tensors with a leading dimension of 1
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Can you specify what errors you get for phi4-mm?

We needed this for sticking with our ckpt format and I remember if we don’t squeeze it will break some deployment flow with TRT-LLM.
It works for all our supported models so far.

# Squeeze scales with a leading dimension of 1
for key, value in post_state_dict.items():
if isinstance(value, torch.Tensor) and value.dim() == 3 and value.shape[0] == 1:
if (
"scale" in key
and isinstance(value, torch.Tensor)
and value.dim() == 3
and value.shape[0] == 1
):
post_state_dict[key] = value.squeeze(0)

# remove real quant parameters from the state dict
keys_to_delete = []
for key, value in post_state_dict.items():
Expand Down
Loading