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

[NPU] chore: bump basic software version to 8.3.rc2 (#14614)

Browse Source
This commit is contained in:
Even Zhou
2025-12-09 09:14:27 +08:00
committed by GitHub
parent 6f657070ef
commit 60d36e7be7
9 changed files with 20 additions and 20 deletions
Download Patch File Download Diff File Expand all files Collapse all files
.github/workflows/nightly-test-npu.yml
+3 -3
View File
@@ -23,7 +23,7 @@ jobs:
matrix: matrix:
part: [0, 1] part: [0, 1]
container: container:
image: swr.cn-southwest-2.myhuaweicloud.com/base_image/ascend-ci/cann:8.3.rc1-a3-ubuntu22.04-py3.11 image: swr.cn-southwest-2.myhuaweicloud.com/base_image/ascend-ci/cann:8.3.rc2-a3-ubuntu22.04-py3.11
steps: steps:
- name: Checkout code - name: Checkout code
uses: actions/checkout@v4 uses: actions/checkout@v4
@@ -69,7 +69,7 @@ jobs:
matrix: matrix:
part: [0] part: [0]
container: container:
image: swr.cn-southwest-2.myhuaweicloud.com/base_image/ascend-ci/cann:8.3.rc1-a3-ubuntu22.04-py3.11 image: swr.cn-southwest-2.myhuaweicloud.com/base_image/ascend-ci/cann:8.3.rc2-a3-ubuntu22.04-py3.11
steps: steps:
- name: Checkout code - name: Checkout code
uses: actions/checkout@v4 uses: actions/checkout@v4
@@ -115,7 +115,7 @@ jobs:
matrix: matrix:
part: [0] part: [0]
container: container:
image: swr.cn-southwest-2.myhuaweicloud.com/base_image/ascend-ci/cann:8.3.rc1-a3-ubuntu22.04-py3.11 image: swr.cn-southwest-2.myhuaweicloud.com/base_image/ascend-ci/cann:8.3.rc2-a3-ubuntu22.04-py3.11
steps: steps:
- name: Checkout code - name: Checkout code
uses: actions/checkout@v4 uses: actions/checkout@v4
.github/workflows/pr-test-npu.yml
+4 -4
View File
@@ -45,7 +45,7 @@ jobs:
if: needs.check-changes.outputs.main_package == 'true' if: needs.check-changes.outputs.main_package == 'true'
runs-on: linux-arm64-npu-1 runs-on: linux-arm64-npu-1
container: container:
image: swr.cn-southwest-2.myhuaweicloud.com/base_image/ascend-ci/cann:8.3.rc1-910b-ubuntu22.04-py3.11 image: swr.cn-southwest-2.myhuaweicloud.com/base_image/ascend-ci/cann:8.3.rc2-910b-ubuntu22.04-py3.11
steps: steps:
- name: Checkout code - name: Checkout code
uses: actions/checkout@v4 uses: actions/checkout@v4
@@ -88,7 +88,7 @@ jobs:
matrix: matrix:
part: [0, 1, 2] part: [0, 1, 2]
container: container:
image: swr.cn-southwest-2.myhuaweicloud.com/base_image/ascend-ci/cann:8.3.rc1-910b-ubuntu22.04-py3.11 image: swr.cn-southwest-2.myhuaweicloud.com/base_image/ascend-ci/cann:8.3.rc2-910b-ubuntu22.04-py3.11
steps: steps:
- name: Checkout code - name: Checkout code
uses: actions/checkout@v4 uses: actions/checkout@v4
@@ -127,7 +127,7 @@ jobs:
if: needs.check-changes.outputs.main_package == 'true' if: needs.check-changes.outputs.main_package == 'true'
runs-on: linux-arm64-npu-4 runs-on: linux-arm64-npu-4
container: container:
image: swr.cn-southwest-2.myhuaweicloud.com/base_image/ascend-ci/cann:8.3.rc1-910b-ubuntu22.04-py3.11 image: swr.cn-southwest-2.myhuaweicloud.com/base_image/ascend-ci/cann:8.3.rc2-910b-ubuntu22.04-py3.11
steps: steps:
- name: Checkout code - name: Checkout code
uses: actions/checkout@v4 uses: actions/checkout@v4
@@ -170,7 +170,7 @@ jobs:
matrix: matrix:
part: [0, 1] part: [0, 1]
container: container:
image: swr.cn-southwest-2.myhuaweicloud.com/base_image/ascend-ci/cann:8.3.rc1-a3-ubuntu22.04-py3.11 image: swr.cn-southwest-2.myhuaweicloud.com/base_image/ascend-ci/cann:8.3.rc2-a3-ubuntu22.04-py3.11
steps: steps:
- name: Checkout code - name: Checkout code
uses: actions/checkout@v4 uses: actions/checkout@v4
.github/workflows/release-docker-npu-nightly.yml
+2 -2
View File
@@ -19,7 +19,7 @@ jobs:
runs-on: ubuntu-22.04-arm runs-on: ubuntu-22.04-arm
strategy: strategy:
matrix: matrix:
cann_version: ["8.3.rc1"] cann_version: ["8.3.rc2"]
device_type: ["910b", "a3"] device_type: ["910b", "a3"]
steps: steps:
- name: Checkout repository - name: Checkout repository
@@ -73,6 +73,6 @@ jobs:
push: ${{ github.repository == 'sgl-project/sglang' && github.event_name != 'pull_request' }} push: ${{ github.repository == 'sgl-project/sglang' && github.event_name != 'pull_request' }}
provenance: false provenance: false
build-args: | build-args: |
SGLANG_KERNEL_NPU_TAG=20251128 SGLANG_KERNEL_NPU_TAG=20251206
CANN_VERSION=${{ matrix.cann_version }} CANN_VERSION=${{ matrix.cann_version }}
DEVICE_TYPE=${{ matrix.device_type }} DEVICE_TYPE=${{ matrix.device_type }}
.github/workflows/release-docker-npu.yml
+2 -2
View File
@@ -17,7 +17,7 @@ jobs:
runs-on: ubuntu-22.04-arm runs-on: ubuntu-22.04-arm
strategy: strategy:
matrix: matrix:
cann_version: ["8.3.rc1"] cann_version: ["8.3.rc2"]
device_type: ["910b", "a3"] device_type: ["910b", "a3"]
steps: steps:
- name: Checkout repository - name: Checkout repository
@@ -70,6 +70,6 @@ jobs:
push: ${{ github.repository == 'sgl-project/sglang' && github.event_name != 'pull_request' }} push: ${{ github.repository == 'sgl-project/sglang' && github.event_name != 'pull_request' }}
provenance: false provenance: false
build-args: | build-args: |
SGLANG_KERNEL_NPU_TAG=20251128 SGLANG_KERNEL_NPU_TAG=20251206
CANN_VERSION=${{ matrix.cann_version }} CANN_VERSION=${{ matrix.cann_version }}
DEVICE_TYPE=${{ matrix.device_type }} DEVICE_TYPE=${{ matrix.device_type }}
docker/npu.Dockerfile
+1 -1
View File
@@ -1,4 +1,4 @@
ARG CANN_VERSION=8.3.rc1 ARG CANN_VERSION=8.3.rc2
ARG DEVICE_TYPE=a3 ARG DEVICE_TYPE=a3
ARG OS=ubuntu22.04 ARG OS=ubuntu22.04
ARG PYTHON_VERSION=py3.11 ARG PYTHON_VERSION=py3.11
docs/platforms/ascend_npu.md
+1 -1
View File
@@ -18,7 +18,7 @@ conda activate sglang_npu
#### CANN #### CANN
Prior to start work with SGLang on Ascend you need to install CANN Toolkit, Kernels operator package and NNAL version 8.3.RC1 or higher, check the [installation guide](https://www.hiascend.com/document/detail/zh/CANNCommunityEdition/83RC1/softwareinst/instg/instg_0008.html?Mode=PmIns&InstallType=local&OS=openEuler&Software=cannToolKit) Prior to start work with SGLang on Ascend you need to install CANN Toolkit, Kernels operator package and NNAL version 8.3.RC2 or higher, check the [installation guide](https://www.hiascend.com/document/detail/zh/CANNCommunityEdition/83RC1/softwareinst/instg/instg_0008.html?Mode=PmIns&InstallType=local&OS=openEuler&Software=cannToolKit)
#### MemFabric Adaptor #### MemFabric Adaptor
docs/platforms/ascend_npu_deepseek_example.md
+5 -5
View File
@@ -30,7 +30,7 @@ python3 -m sglang.launch_server \
--trust-remote-code \ --trust-remote-code \
--attention-backend ascend \ --attention-backend ascend \
--device npu \ --device npu \
--quantization w8a8_int8 \ --quantization modelslim \
--watchdog-timeout 9000 \ --watchdog-timeout 9000 \
--host 127.0.0.1 \ --host 127.0.0.1 \
--port 6688 \ --port 6688 \
@@ -89,7 +89,7 @@ python -m sglang.launch_server \
--mem-fraction-static 0.6 \ --mem-fraction-static 0.6 \
--attention-backend ascend \ --attention-backend ascend \
--device npu \ --device npu \
--quantization w8a8_int8 \ --quantization modelslim \
--disaggregation-transfer-backend ascend \ --disaggregation-transfer-backend ascend \
--max-running-requests 8 \ --max-running-requests 8 \
--context-length 8192 \ --context-length 8192 \
@@ -145,7 +145,7 @@ python -m sglang.launch_server \
--max-running-requests 352 \ --max-running-requests 352 \
--attention-backend ascend \ --attention-backend ascend \
--device npu \ --device npu \
--quantization w8a8_int8 \ --quantization modelslim \
--moe-a2a-backend deepep \ --moe-a2a-backend deepep \
--enable-dp-attention \ --enable-dp-attention \
--deepep-mode low_latency \ --deepep-mode low_latency \
@@ -214,7 +214,7 @@ do
--mem-fraction-static 0.81 \ --mem-fraction-static 0.81 \
--attention-backend ascend \ --attention-backend ascend \
--device npu \ --device npu \
--quantization w8a8_int8 \ --quantization modelslim \
--disaggregation-transfer-backend ascend \ --disaggregation-transfer-backend ascend \
--max-running-requests 8 \ --max-running-requests 8 \
--context-length 8192 \ --context-length 8192 \
@@ -275,7 +275,7 @@ do
--max-running-requests 832 \ --max-running-requests 832 \
--attention-backend ascend \ --attention-backend ascend \
--device npu \ --device npu \
--quantization w8a8_int8 \ --quantization modelslim \
--moe-a2a-backend deepep \ --moe-a2a-backend deepep \
--enable-dp-attention \ --enable-dp-attention \
--deepep-mode low_latency \ --deepep-mode low_latency \
docs/references/mindspore_models.md
+1 -1
View File
@@ -7,7 +7,7 @@ MindSpore is a high-performance AI framework optimized for Ascend NPUs. This doc
## Requirements ## Requirements
MindSpore currently only supports Ascend NPU devices. Users need to first install Ascend CANN software packages. MindSpore currently only supports Ascend NPU devices. Users need to first install Ascend CANN software packages.
The CANN software packages can be downloaded from the [Ascend Official Website](https://www.hiascend.com). The recommended version is 8.3.RC1. The CANN software packages can be downloaded from the [Ascend Official Website](https://www.hiascend.com). The recommended version is 8.3.RC2.
## Supported Models ## Supported Models
scripts/ci/npu_ci_install_dependency.sh
+1 -1
View File
@@ -49,7 +49,7 @@ wget -O "${BISHENG_NAME}" "${BISHENG_URL}" && chmod a+x "${BISHENG_NAME}" && "./
### Install sgl-kernel-npu ### Install sgl-kernel-npu
SGL_KERNEL_NPU_TAG="20251128" SGL_KERNEL_NPU_TAG="20251206"
git clone --depth 1 https://github.com/sgl-project/sgl-kernel-npu.git --branch ${SGL_KERNEL_NPU_TAG} git clone --depth 1 https://github.com/sgl-project/sgl-kernel-npu.git --branch ${SGL_KERNEL_NPU_TAG}
(cd sgl-kernel-npu && bash ./build.sh && ${PIP_INSTALL} output/deep_ep*.whl output/sgl_kernel_npu*.whl && cd "$(python3 -m pip show deep-ep | grep -E '^Location:' | awk '{print $2}')" && ln -s deep_ep/deep_ep_cpp*.so) (cd sgl-kernel-npu && bash ./build.sh && ${PIP_INSTALL} output/deep_ep*.whl output/sgl_kernel_npu*.whl && cd "$(python3 -m pip show deep-ep | grep -E '^Location:' | awk '{print $2}')" && ln -s deep_ep/deep_ep_cpp*.so)