Remove unnecessary load_weights methods (#44589)

Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
This commit is contained in:
Harry Mellor
2026-06-29 01:52:23 -07:00
committed by GitHub
parent db28ae2d07
commit 5051698e41
54 changed files with 821 additions and 1975 deletions
@@ -158,9 +158,6 @@ class Base(
"Transformers modeling backend does "
"not support MXFP4 quantization yet."
)
# Skip loading extra bias for GPTQ models.
if "gptq" in quant_method_name:
self.ignore_unexpected_suffixes.append(".bias")
self._patch_config()
from_config_kwargs = dict(