EngineX/xc-llm-ascend
3
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity

[0.11.0]Chery pick pta upgrade change (#3940)

Browse Source 
This PR cherry-pick two commit from main to upgrade torch-npu to 2.7.1
official release

---------

Signed-off-by: wangxiyuan <wangxiyuan1007@gmail.com>
This commit is contained in:
wangxiyuan
2025-10-31 22:14:26 +08:00
committed by GitHub
parent 3d81ea03ed
commit 8a7154001e
16 changed files with 93 additions and 171 deletions
Download Patch File Download Diff File Expand all files Collapse all files

7
vllm_ascend/ascend_forward_context.py
View File

@@ -11,8 +11,7 @@ from vllm.forward_context import (BatchDescriptor, get_forward_context,
set_forward_context)
import vllm_ascend.envs as envs_ascend
from vllm_ascend.utils import (enable_sp, has_layer_idx, is_moe_model,
version_check)
from vllm_ascend.utils import enable_sp, has_layer_idx, is_moe_model
if TYPE_CHECKING:
from vllm_ascend.ops.weight_prefetch import WeightPrefetchMethod
@@ -162,9 +161,7 @@ def set_ascend_forward_context(
# this optim now just support dense models due to the specific operators used.
# Once the necessary conditions are met, support for MOE models will also be added.
from vllm_ascend.quantization.quant_config import AscendQuantConfig
model_type_scope = ["llama", "qwen2", "qwen3"]
if version_check():
model_type_scope.append("qwen3_moe")
model_type_scope = ["llama", "qwen2", "qwen3", "qwen3_moe"]
addrmsnorm_quant_fusion_enabled = isinstance(vllm_config.quant_config, AscendQuantConfig) and \
vllm_config.model_config.hf_config.model_type in model_type_scope and \
forward_context.layer_idx is not None