[Quant] [Bugfix] Fix quantization config matching with hf_to_vllm_mapper (#20046)

This commit is contained in:
Kyle Sayers
2025-07-01 19:20:34 +09:00
committed by GitHub
parent c05596f1a3
commit 9025a9a705
17 changed files with 107 additions and 29 deletions
@@ -467,6 +467,7 @@ class TransformersForCausalLM(nn.Module, SupportsQuant, SupportsLoRA,
# FIXME(Isotr0py): Don't use any weights mapper for Transformers backend,
# this makes thing complicated. We need to remove this mapper after refactor
# `TransformersModel` in the future.
# NOTE: `SupportsQuant` can be updated after property decorator is removed
@property
def hf_to_vllm_mapper(self):
prefix_mapper = {