Skip to content

Commit cb276cc

Browse files
committed
change for add custom 8bit layer in layer info map
Signed-off-by: unknown <[email protected]>
1 parent da773f1 commit cb276cc

File tree

1 file changed

+1
-1
lines changed

1 file changed

+1
-1
lines changed

modelopt/onnx/quantization/graph_utils.py

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -866,7 +866,7 @@ def get_layer_info(
866866
layers_8bit = kwargs.get("layers_8bit")
867867
gather_block_size = kwargs.get("gather_block_size", DEFAULT_GATHER_BLOCK_SIZE)
868868
gather_quantize_axis = kwargs.get("gather_quantize_axis", DEFAULT_GATHER_QUANTIZE_AXIS)
869-
if enable_mixed_quant:
869+
if enable_mixed_quant or layers_8bit:
870870
layer_info = get_layer_precision_mapping(
871871
onnx_model,
872872
layers_8bit,

0 commit comments

Comments
 (0)