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

update ascend docs (#17987)

Browse Source
This commit is contained in:
amote-i
2026-01-30 18:03:55 +08:00
committed by GitHub
parent 8ce9609fa2
commit 9c2a468e2c
2 changed files with 1 additions and 15 deletions
Download Patch File Download Diff File Expand all files Collapse all files
docs/platforms/ascend_npu.md
-13
View File
@@ -75,19 +75,6 @@ We provide SGL kernels for Ascend NPU, check [installation guide](https://github
#### DeepEP-compatible Library #### DeepEP-compatible Library
We provide a DeepEP-compatible Library as a drop-in replacement of deepseek-ai's DeepEP library, check the [installation guide](https://github.com/sgl-project/sgl-kernel-npu/blob/main/python/deep_ep/README.md). We provide a DeepEP-compatible Library as a drop-in replacement of deepseek-ai's DeepEP library, check the [installation guide](https://github.com/sgl-project/sgl-kernel-npu/blob/main/python/deep_ep/README.md).
#### CustomOps
_TODO: to be removed once merged into sgl-kernel-npu._
Additional package with custom operations. DEVICE_TYPE can be "a3" for Atlas A3 server or "910b" for Atlas A2 server.
```shell
DEVICE_TYPE="a3"
wget https://sglang-ascend.obs.cn-east-3.myhuaweicloud.com/ops/CANN-custom_ops-8.3.0.1-$DEVICE_TYPE-linux.aarch64.run
chmod a+x ./CANN-custom_ops-8.3.0.1-$DEVICE_TYPE-linux.aarch64.run
./CANN-custom_ops-8.3.0.1-$DEVICE_TYPE-linux.aarch64.run --quiet --install-path=/usr/local/Ascend/ascend-toolkit/latest/opp
wget https://sglang-ascend.obs.cn-east-3.myhuaweicloud.com/ops/custom_ops-2.0.$DEVICE_TYPE-cp311-cp311-linux_aarch64.whl
pip install ./custom_ops-2.0.$DEVICE_TYPE-cp311-cp311-linux_aarch64.whl
```
#### Installing SGLang from source #### Installing SGLang from source
```shell ```shell
docs/platforms/ascend_npu_best_practice.md
+1 -2
View File
@@ -936,8 +936,7 @@ export SGLANG_DP_ROUND_ROBIN=1
python -m sglang_router.launch_router \ python -m sglang_router.launch_router \
--pd-disaggregation \ --pd-disaggregation \
--policy cache_aware \ --policy cache_aware \
--prefill http://PIP1:8000 8998 \ --prefill http://PIP1:8000 8995 \
--prefill http://PIP2:8000 8999 \
--decode http://DIP1:8001 \ --decode http://DIP1:8001 \
--host 127.0.0.1 \ --host 127.0.0.1 \
--port 6688 \ --port 6688 \