minke.yu/sglang
1
0
Fork 0
Code Issues Pull Requests Actions 12 Packages Projects Releases Wiki Activity

[NPU][Doc] Update GLM-5 docs, enabling deepep by default (#23708)

Browse Source
This commit is contained in:
cen121212
2026-05-08 11:12:35 +08:00
committed by GitHub
parent 5fa3bb2eaf
commit 461bc8af49
1 changed files with 3 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files
docs/platforms/ascend/ascend_npu_glm5_examples.md
+3 -3
View File
@@ -161,8 +161,6 @@ P_IP=('your ip1' 'your ip2')
P_MASTER="${P_IP[0]}:your port" P_MASTER="${P_IP[0]}:your port"
export SGLANG_DISAGGREGATION_BOOTSTRAP_TIMEOUT=600 export SGLANG_DISAGGREGATION_BOOTSTRAP_TIMEOUT=600
export SGLANG_ENABLE_SPEC_V2=1
export SGLANG_ENABLE_OVERLAP_PLAN_STREAM=1
LOCAL_HOST1=`hostname -I|awk -F " " '{print$1}'` LOCAL_HOST1=`hostname -I|awk -F " " '{print$1}'`
LOCAL_HOST2=`hostname -I|awk -F " " '{print$2}'` LOCAL_HOST2=`hostname -I|awk -F " " '{print$2}'`
@@ -182,7 +180,9 @@ do
--mem-fraction-static 0.8\ --mem-fraction-static 0.8\
--port 8000 \ --port 8000 \
--served-model-name glm-5 \ --served-model-name glm-5 \
--cuda-graph-max-bs 16 \ --cuda-graph-max-bs 32 \
--moe-a2a-backend deepep \
--deepep-mode auto \
--disable-radix-cache --disable-radix-cache
NODE_RANK=$i NODE_RANK=$i
break break