[Quant] Fix use_mla TypeError and support loading pure-sparsity Compressed Tensors configs (#12711)

This commit is contained in:
Kyle Sayers
2025-02-03 23:27:11 -08:00
committed by GitHub
parent bb392af434
commit 4896d0c2dd
2 changed files with 8 additions and 2 deletions
@@ -424,6 +424,11 @@ class CompressedTensorsConfig(QuantizationConfig):
or input_quant is not None,
weight_quant=weight_quant,
input_quant=input_quant)
elif weight_quant is None:
logger.warning_once("Acceleration for non-quantized schemes is "
"not supported by Compressed Tensors. "
"Falling back to UnquantizedLinearMethod")
return None
else:
# Find the quant_scheme
scheme = self._get_scheme_from_parts( # type: ignore