CookieApril/enginex-ascend-910-vllm
1
0
Fork 0
forked from EngineX-Ascend/enginex-ascend-910-vllm
Code Pull Requests Activity

init v0.11.0rc0

Browse Source
This commit is contained in:
luopingyi
2025-10-14 10:38:28 +08:00
parent 67afd0ea78
commit 66dc16f966
278 changed files with 28130 additions and 11708 deletions
Download Patch File Download Diff File Expand all files Collapse all files

7
vllm_ascend/torchair/models/qwen2.py
View File

@@ -40,7 +40,6 @@ from vllm.model_executor.models.qwen2 import Qwen2ForCausalLM # noqa: F401
from vllm.model_executor.models.qwen2 import Qwen2MLP, Qwen2Model
from vllm.model_executor.models.utils import (AutoWeightsLoader,
PPMissingLayer, maybe_prefix)
from vllm.model_executor.sampling_metadata import SamplingMetadata
from vllm.sequence import IntermediateTensors
from vllm_ascend.ascend_config import get_ascend_config
@@ -343,9 +342,9 @@ class CustomQwen2ForCausalLM(nn.Module, SupportsLoRA, SupportsPP):
return hidden_states
def compute_logits(
self,
hidden_states: torch.Tensor,
sampling_metadata: SamplingMetadata,
self,
hidden_states: torch.Tensor,
sampling_metadata=None, # type: ignore
) -> Optional[torch.Tensor]:
logits = self.logits_processor(self.lm_head, hidden_states,
sampling_metadata)