apps/FastDeploy
1
0
Fork 0
mirror of https://github.com/PaddlePaddle/FastDeploy.git synced 2025-10-04 00:06:38 +08:00
Code Issues Actions 2 Packages Projects Releases Wiki Activity

[feat] support fa3 backend for pd disaggregated (#2695)
Some checks failed
Deploy GitHub Pages / deploy (push) Has been cancelled
Details

Browse Source 
* support fa3 backend run in pd disaggregated

* support fa3 backend run in pd disaggregated

* support fa3 backend run in pd disaggregated

* support fa3 backend run in pd disaggregated

* delete use_fast_ffn
This commit is contained in:
Yuanle Liu
2025-07-03 22:33:27 +08:00
committed by GitHub
parent 00863c43fd
commit 240bdac2a4
26 changed files with 455 additions and 139 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
fastdeploy/model_executor/models/qwen2.py
View File

@@ -27,7 +27,7 @@ from fastdeploy.config import FDConfig, ModelConfig
from fastdeploy.model_executor.graph_optimization.decorator import \
support_graph_optimization
from fastdeploy.model_executor.layers.activation import SiluAndMul
from fastdeploy.model_executor.layers.attention import Attention
from fastdeploy.model_executor.layers.attention.attention import Attention
from fastdeploy.model_executor.layers.embeddings import VocabParallelEmbedding
from fastdeploy.model_executor.layers.linear import (
MergedColumnParallelLinear, QKVParallelLinear, RowParallelLinear)
@@ -55,7 +55,6 @@ class Qwen2MLP(nn.Layer):
output_size=fd_config.model_config.ffn_hidden_size * 2,
with_bias=False,
activation=fd_config.model_config.hidden_act,
use_fast_ffn=True,
)
self.down_proj = RowParallelLinear(