diff --git a/vllm/model_executor/models/deepseek_v2.py b/vllm/model_executor/models/deepseek_v2.py index 38114836bfdbb..d4ad0c6b5c99e 100644 --- a/vllm/model_executor/models/deepseek_v2.py +++ b/vllm/model_executor/models/deepseek_v2.py @@ -28,6 +28,7 @@ from transformers import PretrainedConfig from vllm.attention import Attention, AttentionMetadata +from vllm.compilation.decorators import support_torch_compile from vllm.config import CacheConfig from vllm.distributed import (get_pp_group, get_tensor_model_parallel_world_size, @@ -403,6 +404,7 @@ def forward( return hidden_states, residual +@support_torch_compile class DeepseekV2Model(nn.Module): fall_back_to_pt_during_load = False