ayo/ipex-llm
1
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions 1

Fix vllm glm edge model (#13007)

Browse source 
* fix done

* fix
This commit is contained in:
Guancheng Fu 2025-03-26 09:25:32 +08:00 committed by GitHub
parent 374747b492
commit f437b36678
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
1 changed files with 2 additions and 1 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

3
python/llm/src/ipex_llm/transformers/convert.py
View file

@ -192,7 +192,8 @@ def is_linear_module(module):
and hasattr(module.quant_method, "quant_config") and hasattr(module.quant_method, "quant_config")
and module.quant_method.quant_config.get_name() == "gptq"): and module.quant_method.quant_config.get_name() == "gptq"):
_USE_VLLM_GPTQ = True _USE_VLLM_GPTQ = True
invalidInputError(module.skip_bias_add is not True, "Currently, ipex-vllm does not" invalidInputError(module.skip_bias_add is not True or module.bias is None,
"Currently, ipex-vllm does not"
" support linear layers with skip_bias_add argument") " support linear layers with skip_bias_add argument")
if isinstance(module, RowParallelLinear) and tp_size >= 2: if isinstance(module, RowParallelLinear) and tp_size >= 2:
mp_group = get_tensor_model_parallel_group() mp_group = get_tensor_model_parallel_group()