Skip to content

fix rtn bug #1622

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
wenhuach21 merged 2 commits into from
Mar 26, 2026
Merged

fix rtn bug #1622

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
e90c924
fix
wenhuach21 Mar 26, 2026
6aae77e
[pre-commit.ci] auto fixes from pre-commit.com hooks
pre-commit-ci[bot] Mar 26, 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
4 changes: 3 additions & 1 deletion auto_round/compressors/base.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1316,7 +1316,9 @@ def _immediate_pack(self, name: str):
tokenizer=self.tokenizer,
)

@torch.inference_mode()
# Use no_grad instead of inference mode
# https://github.com/intel/auto-round/issues/1620
@torch.no_grad()
def _quantize_rtn(self) -> tuple[torch.nn.Module, dict[str, Any]]:
"""Quantize all modules in the model using RTN (Round-To-Nearest) strategy.

Expand Down
2 changes: 1 addition & 1 deletion auto_round/utils/common.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -464,7 +464,7 @@ def __getitem__(self, key):
if importlib.util.find_spec("deepspeed"): # check if deepspeed is installed
deepspeed_exists = True

SUPPORTED_DTYPES = ("int", "mx_fp", "fp", "nv_fp")
SUPPORTED_DTYPES = ("int", "mx_fp", "fp", "nv_fp", "mx_int")
SUPPORTED_FORMATS = SupportedFormats()
SUPPORTED_LAYER_TYPES = (torch.nn.Linear, transformers.pytorch_utils.Conv1D)
# Changed to str as it relies on triton or others lib to load this
Expand Down
Loading