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

19
vllm_ascend/torchair/models/qwen3_moe.py
View File

@@ -54,8 +54,9 @@ from vllm.sequence import IntermediateTensors
from vllm_ascend.ascend_config import get_ascend_config
from vllm_ascend.attention.attention_v1 import AscendAttentionState
from vllm_ascend.ops.fused_moe import AscendFusedMoE
from vllm_ascend.ops.sequence_parallel import (MetadataForPadding,
init_metadata_for_sp)
from vllm_ascend.torchair.ops.sequence_parallel import (MetadataForPadding,
init_metadata_for_sp)
from vllm_ascend.utils import vllm_version_is
class CustomSparseMoeBlock(Qwen3MoeSparseMoeBlock):
@@ -311,9 +312,14 @@ class CustomQwen3MoeDecoderLayer(Qwen3MoeDecoderLayer):
quant_config=quant_config,
prefix=f"{prefix}.mlp")
else:
self.mlp = Qwen3MoeSparseMoeBlock(config=config,
quant_config=quant_config,
prefix=f"{prefix}.mlp")
if vllm_version_is("0.10.2"):
self.mlp = Qwen3MoeSparseMoeBlock(
config=config,
quant_config=quant_config,
prefix=f"{prefix}.mlp")
else:
self.mlp = Qwen3MoeSparseMoeBlock(vllm_config=vllm_config,
prefix=f"{prefix}.mlp")
else:
self.mlp = Qwen3MoeMLP(hidden_size=config.hidden_size,
intermediate_size=config.intermediate_size,
@@ -394,7 +400,8 @@ class CustomQwen3MoeModel(Qwen3MoeModel):
quant_config = vllm_config.quant_config
parallel_config = vllm_config.parallel_config
self.num_redundant_experts = parallel_config.num_redundant_experts
eplb_config = parallel_config.eplb_config
self.num_redundant_experts = eplb_config.num_redundant_experts
self.padding_idx = config.pad_token_id
self.vocab_size = config.vocab_size
self.config = config