Skip to content
Open
Show file tree
Hide file tree
Changes from all commits
Commits
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
3 changes: 1 addition & 2 deletions setup.py
Original file line number Diff line number Diff line change
Expand Up @@ -144,8 +144,7 @@ def localversion_func(version: ScmVersion) -> str:
if BUILD_TYPE == "release"
else "compressed-tensors>=0.12.3a2"
),
# TODO: replace it with the release version
("auto_round @ git+https://github.com/intel/auto-round.git@llmc"),
("auto-round==0.9.1"),
],
extras_require={
"dev": [
Expand Down
4 changes: 3 additions & 1 deletion src/llmcompressor/modifiers/autoround/base.py
Original file line number Diff line number Diff line change
Expand Up @@ -107,6 +107,7 @@ class AutoRoundModifier(Modifier, QuantizationMixin):
# AutoRound modifier arguments
iters: int = 200
enable_torch_compile: bool = True
batch_size: int = 8

# private variables
_module_names: Dict[torch.nn.Module, str] = PrivateAttr(default_factory=dict)
Expand Down Expand Up @@ -223,6 +224,7 @@ def apply_autoround(self, state, subgraph):
scheme=ar_quant_scheme,
iters=self.iters,
enable_torch_compile=self.enable_torch_compile,
batch_size=self.batch_size,
)
# TODO: configure layer-wise config based on self.resolved_config
ar.configure_layer_config(enable_gguf_official_mixed=False)
Expand All @@ -236,7 +238,7 @@ def apply_autoround(self, state, subgraph):
block=decoding_layer,
inputs=cur_inputs,
q_input=self._q_input,
device=device,
device=str(device),
# Leave offload for LLMC
auto_offload=False,
)
Expand Down