EngineX-Hygon/sglang
5
0
Fork 0
Code Issues Pull Requests Actions 7 Projects Releases Wiki Activity

[CPU] fix CPU backend sel. issue for Llama4 (#10511)

Browse Source
This commit is contained in:
Zaili Wang
2025-09-16 17:57:45 +08:00
committed by GitHub
parent 8df7353af3
commit 925dbb3218
2 changed files with 3 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
docs/platforms/cpu_server.md
View File

@@ -81,6 +81,8 @@ git clone https://github.com/sgl-project/sglang.git
cd sglang cd sglang
git checkout <YOUR-DESIRED-VERSION> git checkout <YOUR-DESIRED-VERSION>
# Use dedicated toml file
cp python/pyproject_other.toml python/pyproject.toml
# Install SGLang dependent libs, and build SGLang main package # Install SGLang dependent libs, and build SGLang main package
pip install --upgrade pip setuptools pip install --upgrade pip setuptools
conda install -y libsqlite==3.48.0 gperftools tbb libnuma numactl conda install -y libsqlite==3.48.0 gperftools tbb libnuma numactl

2
python/sglang/srt/server_args.py
View File

@@ -2648,7 +2648,7 @@ class ServerArgs:
# use bf16 for mxfp4 triton kernels # use bf16 for mxfp4 triton kernels
self.dtype = "bfloat16" self.dtype = "bfloat16"
elif "Llama4" in model_arch: elif "Llama4" in model_arch and self.device != "cpu":
assert self.attention_backend in { assert self.attention_backend in {
"fa3", "fa3",
"aiter", "aiter",