Skip to content

fix gptq quantization condition #2416

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
IlyasMoutawwakil merged 2 commits into from
Apr 27, 2026
Merged

fix gptq quantization condition #2416

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
218ec23
fix gptq quantization condition
jiqing-feng Mar 25, 2026
5df216c
revert useless change
jiqing-feng Mar 25, 2026
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
10 changes: 9 additions & 1 deletion optimum/gptq/utils.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -47,7 +47,15 @@ def get_layers(module: nn.Module, layers=[Conv1D, nn.Conv2d, nn.Linear], prefix:
`Dict[str,Union[Conv1D, nn.Conv2d, nn.Linear]]`: Mapping of the name of the layer and the actual layer
"""
for layer in layers:
if isinstance(module, layer):
# Use exact type check for nn.Linear to skip subclasses with custom forward (e.g., MoE routers)
if layer is nn.Linear:
if type(module) is nn.Linear:
if prefix is not None:
if name.startswith(prefix):
return {name: module}
else:
return {name: module}
elif isinstance(module, layer):
if prefix is not None:
if name.startswith(prefix):
return {name: module}
Expand Down