Back to Sglang

SGLang installation with NPUs support

docs_new/docs/hardware-platforms/ascend-npus/ascend_npu.mdx

0.5.1318.7 KB
Original Source

You can install SGLang using any of the methods below. Please go through System Settings section to ensure the clusters are roaring at max performance. Feel free to leave an issue here at sglang if you encounter any issues or have any problems.

Component Version Mapping For SGLang

<table style={{width: "100%", borderCollapse: "collapse", tableLayout: "fixed"}}> <colgroup> <col style={{width: "34%"}} /> <col style={{width: "33%"}} /> <col style={{width: "33%"}} /> </colgroup> <thead> <tr style={{borderBottom: "2px solid #d55816"}}> <th style={{textAlign: "left", padding: "10px 12px", fontWeight: 700, whiteSpace: "nowrap", backgroundColor: "rgba(255,255,255,0.02)"}}>Component</th> <th style={{textAlign: "left", padding: "10px 12px", fontWeight: 700, whiteSpace: "nowrap", backgroundColor: "rgba(255,255,255,0.05)"}}>Version</th> <th style={{textAlign: "left", padding: "10px 12px", fontWeight: 700, whiteSpace: "nowrap", backgroundColor: "rgba(255,255,255,0.02)"}}>Obtain Way</th> </tr> </thead> <tbody> <tr> <td style={{padding: "9px 12px", fontWeight: 500, backgroundColor: "rgba(255,255,255,0.02)"}}>HDK</td> <td style={{padding: "9px 12px", backgroundColor: "rgba(255,255,255,0.05)"}}>26.0.RC1</td> <td style={{padding: "9px 12px", backgroundColor: "rgba(255,255,255,0.02)"}}><a href="https://www.hiascend.com/hardware/firmware-drivers/community?product=1&model=30&cann=9.0.0&driver=Ascend+HDK+26.0.RC1">link</a></td> </tr> <tr> <td style={{padding: "9px 12px", fontWeight: 500, backgroundColor: "rgba(255,255,255,0.02)"}}>CANN</td> <td style={{padding: "9px 12px", backgroundColor: "rgba(255,255,255,0.05)"}}>9.0.0</td> <td style={{padding: "9px 12px", backgroundColor: "rgba(255,255,255,0.02)"}}><a href="#obtain-cann-image">Obtain Images</a></td> </tr> <tr> <td style={{padding: "9px 12px", fontWeight: 500, backgroundColor: "rgba(255,255,255,0.02)"}}>Pytorch Adapter</td> <td style={{padding: "9px 12px", backgroundColor: "rgba(255,255,255,0.05)"}}>26.0.0</td> <td style={{padding: "9px 12px", backgroundColor: "rgba(255,255,255,0.02)"}}><a href="https://gitcode.com/Ascend/pytorch/releases">link</a></td> </tr> <tr> <td style={{padding: "9px 12px", fontWeight: 500, backgroundColor: "rgba(255,255,255,0.02)"}}>MemFabric</td> <td style={{padding: "9px 12px", backgroundColor: "rgba(255,255,255,0.05)"}}>1.0.8</td> <td style={{padding: "9px 12px", backgroundColor: "rgba(255,255,255,0.02)"}}>`pip install memfabric-hybrid==1.0.8`</td> </tr> <tr> <td style={{padding: "9px 12px", fontWeight: 500, backgroundColor: "rgba(255,255,255,0.02)"}}>Triton</td> <td style={{padding: "9px 12px", backgroundColor: "rgba(255,255,255,0.05)"}}>3.2.1</td> <td style={{padding: "9px 12px", backgroundColor: "rgba(255,255,255,0.02)"}}><a href="https://gitcode.com/Ascend/triton-ascend/releases/v3.2.1">link</a></td> </tr> <tr> <td style={{padding: "9px 12px", fontWeight: 500, backgroundColor: "rgba(255,255,255,0.02)"}}>SGLang NPU Kernel</td> <td style={{padding: "9px 12px", backgroundColor: "rgba(255,255,255,0.05)"}}>2026.05.01.post3</td> <td style={{padding: "9px 12px", backgroundColor: "rgba(255,255,255,0.02)"}}><a href="https://github.com/sgl-project/sgl-kernel-npu/releases/tag/2026.05.01.post3">link</a></td> </tr> <tr> <td style={{padding: "9px 12px", fontWeight: 500, backgroundColor: "rgba(255,255,255,0.02)"}}>MemFabric-zbal</td> <td style={{padding: "9px 12px", backgroundColor: "rgba(255,255,255,0.05)"}}>1.1.1</td> <td style={{padding: "9px 12px", backgroundColor: "rgba(255,255,255,0.02)"}}>`pip install memfabric-zbal==1.1.1`</td> </tr> </tbody> </table>

Obtain CANN Image

<Warning> Ensure sufficient disk space before pulling images. Each Docker image requires at least **30 GB** of free space. </Warning>

You can obtain the dependency of a specified version of CANN through an image.

<Tabs> <Tab title="Atlas 800I A3">
bash
docker pull quay.io/ascend/cann:9.0.0-a3-ubuntu22.04-py3.11
</Tab> <Tab title="Atlas 800I A2">
bash
docker pull quay.io/ascend/cann:9.0.0-910b-ubuntu22.04-py3.11
</Tab> </Tabs>

Preparing the Running Environment

Method 1: Installing from source with prerequisites

Python Version

Only python==3.11 is supported currently. If you don't want to break system pre-installed python, try installing with conda.

bash
conda create --name sglang_npu python=3.11
conda activate sglang_npu

Note on Anaconda repository restrictions If you encounter an error like “Terms of Service have not been accepted” during the conda create step, the default Anaconda repository is blocking package downloads. To resolve this, configure a mirror (e.g., Tsinghua Open Source Mirror):

bash
# Add Tsinghua mirrors
conda config --add channels https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/main/
conda config --add channels https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/conda-forge/
conda config --set show_channel_urls yes

# Edit the system-level conda config to remove any hardcoded defaults
vi /root/miniconda3/.condarc

Inside /root/miniconda3/.condarc, delete or comment out any lines containing defaults or official Anaconda URLs. Then remove the failed environment and recreate it:

bash
conda clean -i
conda env remove -n sglang_npu
conda create --name sglang_npu python=3.11
conda activate sglang_npu

CANN

Prior to start work with SGLang on Ascend you need to install CANN Toolkit, Kernels operator package and NNAL version 9.0.0, check the installation guide

MemFabric-Hybrid

If you want to use PD disaggregation mode, you need to install MemFabric-Hybrid. MemFabric-Hybrid is a drop-in replacement of Mooncake Transfer Engine that enables KV cache transfer on Ascend NPU clusters.

bash
pip install memfabric-hybrid==1.0.5

Pytorch and Pytorch Framework Adaptor on Ascend

bash
PYTORCH_VERSION=2.8.0
TORCHVISION_VERSION=0.23.0
TORCH_NPU_VERSION=2.8.0.post2
pip install torch==$PYTORCH_VERSION torchvision==$TORCHVISION_VERSION --index-url https://download.pytorch.org/whl/cpu
pip install torch_npu==$TORCH_NPU_VERSION

If you are using other versions of torch and install torch_npu, check installation guide

Triton on Ascend

We provide our own implementation of Triton for Ascend.

bash
pip install triton-ascend

For installation of Triton on Ascend nightly builds or from sources, follow installation guide

SGLang Kernels NPU

We provide SGL kernels for Ascend NPU, check installation guide.

DeepEP-compatible Library

We provide a DeepEP-compatible Library as a drop-in replacement of deepseek-ai's DeepEP library, check the installation guide.

Some other dependencies

bash
# libGL
apt update
apt install libgl1 libglib2.0-0

# ensure setuptools contains pkg_resources module
pip install "setuptools<80"

Installing SGLang from source

bash
# Use the last release branch
git clone https://github.com/sgl-project/sglang.git
cd sglang
mv python/pyproject_npu.toml python/pyproject.toml
pip install -e python[all_npu]

Method 2: Using Docker Image

Obtain Image

You can download the SGLang image or build an image based on Dockerfile to obtain the Ascend NPU image.

<Warning> Ensure sufficient disk space before pulling images. Each Docker image requires at least **30 GB** of free space. If you need to download model weights, check the model size at [ModelScope](https://www.modelscope.cn/models) to reserve enough space. </Warning>
  1. Download SGLang image
<Tip> We publish both **stable releases** and **daily builds**. Choose a stable release tag (e.g., `v0.5.10-npu.rc1-a3`) if you prefer a validated version, or a daily build tag (e.g., `main-cann9.0.0-a3`) if you need the latest development changes. </Tip> <Tabs> <Tab title="Atlas 800I A3">
bash
# Stable release
docker pull quay.io/ascend/sglang:v0.5.10-npu.rc1-a3

# Daily build
docker pull quay.io/ascend/sglang:main-cann9.0.0-a3
</Tab> <Tab title="Atlas 800I A2">
bash
# Stable release
docker pull quay.io/ascend/sglang:v0.5.10-npu.rc1-910b

# Daily build
docker pull quay.io/ascend/sglang:main-cann9.0.0-910b
</Tab> </Tabs>
  1. Build an image based on Dockerfile
bash
# Clone the SGLang repository
git clone https://github.com/sgl-project/sglang.git
cd sglang/docker

# Build the docker image
# If there are network errors, please modify the Dockerfile to use offline dependencies or use a proxy
# <arch_tag> is the target architecture of the image, e.g. amd64, arm64
docker build --build-arg TARGETARCH=<arch_tag> -t <image_name> -f npu.Dockerfile .

Create Docker

Notice: --privileged and --network=host are required by RDMA, which is typically needed by Ascend NPU clusters.

<Tabs> <Tab title="Atlas 800I A3">
bash
alias drun='docker run -it --rm --privileged --network=host --ipc=host --shm-size=16g \
    --device=/dev/davinci0 --device=/dev/davinci1 --device=/dev/davinci2 --device=/dev/davinci3 \
    --device=/dev/davinci4 --device=/dev/davinci5 --device=/dev/davinci6 --device=/dev/davinci7 \
    --device=/dev/davinci8 --device=/dev/davinci9 --device=/dev/davinci10 --device=/dev/davinci11 \
    --device=/dev/davinci12 --device=/dev/davinci13 --device=/dev/davinci14 --device=/dev/davinci15 \
    --device=/dev/davinci_manager --device=/dev/hisi_hdc \
    --volume /usr/local/sbin:/usr/local/sbin --volume /usr/local/Ascend/driver:/usr/local/Ascend/driver \
    --volume /usr/local/Ascend/firmware:/usr/local/Ascend/firmware \
    --volume /etc/ascend_install.info:/etc/ascend_install.info \
    --volume /var/queue_schedule:/var/queue_schedule --volume ~/.cache/:/root/.cache/'

# Add HF_TOKEN env for download model by SGLang.
drun --env "HF_TOKEN=<secret>" \
    <image_name> \
    python3 -m sglang.launch_server --model-path meta-llama/Llama-3.1-8B-Instruct --attention-backend ascend
</Tab> <Tab title="Atlas 800I A2">
bash
alias drun='docker run -it --rm --privileged --network=host --ipc=host --shm-size=16g \
    --device=/dev/davinci0 --device=/dev/davinci1 --device=/dev/davinci2 --device=/dev/davinci3 \
    --device=/dev/davinci4 --device=/dev/davinci5 --device=/dev/davinci6 --device=/dev/davinci7 \
    --device=/dev/davinci_manager --device=/dev/hisi_hdc \
    --volume /usr/local/sbin:/usr/local/sbin --volume /usr/local/Ascend/driver:/usr/local/Ascend/driver \
    --volume /usr/local/Ascend/firmware:/usr/local/Ascend/firmware \
    --volume /etc/ascend_install.info:/etc/ascend_install.info \
    --volume /var/queue_schedule:/var/queue_schedule --volume ~/.cache/:/root/.cache/'

# Add HF_TOKEN env for download model by SGLang.
drun --env "HF_TOKEN=<secret>" \
    <image_name> \
    python3 -m sglang.launch_server --model-path meta-llama/Llama-3.1-8B-Instruct --attention-backend ascend
</Tab> </Tabs> <Tip> SGLang will serve on `http://127.0.0.1:30000` by default. You can change the host and port by `--host` and `--port` parameters. </Tip>

System Settings

CPU performance power scheme

The default power scheme on Ascend hardware is ondemand which could affect performance, changing it to performance is recommended.

bash
echo performance | sudo tee /sys/devices/system/cpu/cpu*/cpufreq/scaling_governor

# Make sure changes are applied successfully
cat /sys/devices/system/cpu/cpu0/cpufreq/scaling_governor # shows performance

Disable NUMA balancing

bash
sudo sysctl -w kernel.numa_balancing=0
# Check
cat /proc/sys/kernel/numa_balancing # shows 0

Prevent swapping out system memory

bash
sudo sysctl -w vm.swappiness=10

# Check
cat /proc/sys/vm/swappiness # shows 10

Running SGLang Service

Running Service For Large Language Models

PD Mixed Scene

bash
# Enabling CPU Affinity
export SGLANG_SET_CPU_AFFINITY=1
python3 -m sglang.launch_server \
  --model-path meta-llama/Llama-3.1-8B-Instruct \
  --attention-backend ascend \
  --host 127.0.0.1 \
  --port 8000

PD Disaggregation Scene

  1. Launch Prefill Server
<Tabs> <Tab title="Atlas 800I A3">
bash
# Enabling CPU Affinity
export SGLANG_SET_CPU_AFFINITY=1

# PREFILL_IP: IP address of the first Prefill Server
# FREE_PORT: any available port
# all SGLang servers need to be configured with the same PREFILL_IP and FREE_PORT
export ASCEND_MF_STORE_URL="tcp://PREFILL_IP:FREE_PORT"
python3 -m sglang.launch_server \
    --model-path meta-llama/Llama-3.1-8B-Instruct \
    --disaggregation-mode prefill \
    --disaggregation-transfer-backend ascend \
    --disaggregation-bootstrap-port 8995 \
    --attention-backend ascend \
    --device npu \
    --base-gpu-id 0 \
    --tp-size 1 \
    --host 127.0.0.1 \
    --port 8000
</Tab> <Tab title="Atlas 800I A2">
bash
# Enabling CPU Affinity
export SGLANG_SET_CPU_AFFINITY=1

# PREFILL_IP: IP address of the first Prefill Server
# FREE_PORT: any available port
# all SGLang servers need to be configured with the same PREFILL_IP and FREE_PORT
export ASCEND_MF_STORE_URL="tcp://PREFILL_IP:FREE_PORT"
export ASCEND_MF_TRANSFER_PROTOCOL="device_rdma"
python3 -m sglang.launch_server \
    --model-path meta-llama/Llama-3.1-8B-Instruct \
    --disaggregation-mode prefill \
    --disaggregation-transfer-backend ascend \
    --disaggregation-bootstrap-port 8995 \
    --attention-backend ascend \
    --device npu \
    --base-gpu-id 0 \
    --tp-size 1 \
    --host 127.0.0.1 \
    --port 8000
</Tab> </Tabs>
  1. Launch Decode Server
<Tabs> <Tab title="Atlas 800I A3">
bash
# PREFILL_IP: IP address of the first Prefill Server
# FREE_PORT: any available port
# all SGLang servers need to be configured with the same PREFILL_IP and FREE_PORT
export ASCEND_MF_STORE_URL="tcp://PREFILL_IP:FREE_PORT"
python3 -m sglang.launch_server \
    --model-path meta-llama/Llama-3.1-8B-Instruct \
    --disaggregation-mode decode \
    --disaggregation-transfer-backend ascend \
    --attention-backend ascend \
    --device npu \
    --base-gpu-id 1 \
    --tp-size 1 \
    --host 127.0.0.1 \
    --port 8001
</Tab> <Tab title="Atlas 800I A2">
bash
# PREFILL_IP: IP address of the first Prefill Server
# FREE_PORT: any available port
# all SGLang servers need to be configured with the same PREFILL_IP and FREE_PORT
export ASCEND_MF_STORE_URL="tcp://PREFILL_IP:FREE_PORT"
export ASCEND_MF_TRANSFER_PROTOCOL="device_rdma"
python3 -m sglang.launch_server \
    --model-path meta-llama/Llama-3.1-8B-Instruct \
    --disaggregation-mode decode \
    --disaggregation-transfer-backend ascend \
    --attention-backend ascend \
    --device npu \
    --base-gpu-id 1 \
    --tp-size 1 \
    --host 127.0.0.1 \
    --port 8001
</Tab> </Tabs>
  1. Launch Router
bash
python3 -m sglang_router.launch_router \
    --pd-disaggregation \
    --policy cache_aware \
    --prefill http://127.0.0.1:8000 8995 \
    --decode http://127.0.0.1:8001 \
    --host 127.0.0.1 \
    --port 6688

Running Service For Multimodal Language Models

PD Mixed Scene

bash
python3 -m sglang.launch_server \
    --model-path Qwen3-VL-30B-A3B-Instruct \
    --host 127.0.0.1 \
    --port 8000 \
    --tp 4 \
    --device npu \
    --attention-backend ascend \
    --mm-attention-backend ascend_attn \
    --disable-radix-cache \
    --trust-remote-code \
    --enable-multimodal \
    --sampling-backend ascend

Testing the Service

Once the server prints The server is fired up and ready to roll! in the logs, it is ready to accept requests.

Which port to send requests to

The port you use depends on your deployment mode:

ScenarioWhere to send requests
Non-PD (single server)The server's --port (e.g., 8000 in the examples above)
Non-PD (multi-node)The primary node's (--node-rank 0) --port; do not send requests to worker nodes
PD disaggregationThe router's --port (e.g., 6688 in the examples above); do not send requests directly to prefill or decode servers
<Tip> SGLang serves on port `30000` by default if `--port` is not specified. The examples in this guide use explicit ports for clarity.

If you are using PD disaggregation, replace 8000 with your router's port (e.g., 6688) in the following examples. </Tip>

Health Check

bash
curl http://127.0.0.1:8000/health

A successful response returns HTTP 200 with an empty body.

Generate (Native Endpoint)

bash
curl http://127.0.0.1:8000/generate \
  -H "Content-Type: application/json" \
  -d '{
    "text": "What is the capital of France?",
    "sampling_params": {"temperature": 0, "max_new_tokens": 128}
  }'

The expected output should contain "Paris".

Chat Completions (OpenAI-Compatible)

bash
curl http://127.0.0.1:8000/v1/chat/completions \
  -H "Content-Type: application/json" \
  -d '{
    "model": "meta-llama/Llama-3.1-8B-Instruct",
    "messages": [{"role": "user", "content": "What is the capital of France?"}]
  }'

Some models return responses accompanied with thinking process content. To disable this output, configure parameters as follows:

bash
curl http://127.0.0.1:8000/v1/chat/completions
  -H "Content-Type: application/json"
  -d '{
    "model": "Eco-Tech/Qwen3.5-27B-w8a8-mtp",
    "messages": [{"role": "user", "content": "What is the capital of France?"}],
    "chat_template_kwargs": {"enable_thinking": false}
  }'

The expected output should contain "Paris".

Multimodal Chat Completions

<Note> The image URL in the example below references an external resource (`raw.githubusercontent.com`). Make sure the server has internet access so the image can be downloaded at inference time. Alternatively, you can use a locally accessible URL or base64-encoded image data. </Note>
bash
curl http://127.0.0.1:8000/v1/chat/completions \
  -H "Content-Type: application/json" \
  -d '{
    "model": "Qwen3-VL-30B-A3B-Instruct",
    "messages": [{
      "role": "user",
      "content": [
        {"type": "image_url", "image_url": {"url": "https://raw.githubusercontent.com/sgl-project/sglang/main/examples/assets/example_image.png"}},
        {"type": "text", "text": "Describe this image."}
      ]
    }]
  }'