biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

Separate base model from TransformersModel (#15467)

Browse Source 
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
Signed-off-by: Isotr0py <2037008807@qq.com>
Co-authored-by: Isotr0py <2037008807@qq.com>
Co-authored-by: Isotr0py <mozf@mail2.sysu.edu.cn>
This commit is contained in:
Harry Mellor
2025-03-26 10:13:38 +00:00
committed by GitHub
parent 4ec2cee000
commit cf5c8f1686
6 changed files with 110 additions and 59 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
vllm/model_executor/model_loader/utils.py
View File

@@ -45,7 +45,7 @@ def is_transformers_impl_compatible(
def resolve_transformers_fallback(model_config: ModelConfig,
architectures: list[str]):
for i, arch in enumerate(architectures):
if arch == "TransformersModel":
if arch == "TransformersForCausalLM":
continue
auto_map: dict[str, str] = getattr(model_config.hf_config, "auto_map",
None) or dict()
@@ -69,7 +69,7 @@ def resolve_transformers_fallback(model_config: ModelConfig,
raise ValueError(
f"The Transformers implementation of {arch} is not "
"compatible with vLLM.")
architectures[i] = "TransformersModel"
architectures[i] = "TransformersForCausalLM"
if model_config.model_impl == ModelImpl.AUTO:
if not is_transformers_impl_compatible(arch, custom_model_module):
raise ValueError(
@@ -80,7 +80,7 @@ def resolve_transformers_fallback(model_config: ModelConfig,
"%s has no vLLM implementation, falling back to Transformers "
"implementation. Some features may not be supported and "
"performance may not be optimal.", arch)
architectures[i] = "TransformersModel"
architectures[i] = "TransformersForCausalLM"
return architectures