Skip to content

Commit 4bdec7e

Browse files
zhaochen20zhaochen20
committed
Update 2025-03-13 10:06:27
1 parent ade958b commit 4bdec7e

File tree

71 files changed

+4569
-4553
lines changed

Some content is hidden

Large Commits have some content hidden by default. Use the searchbox below for content that may be hidden.

71 files changed

+4569
-4553
lines changed

‎README.html

+2-2
Original file line numberDiff line numberDiff line change
@@ -42,7 +42,7 @@
4242
<link rel="preload" as="script" href="_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
4343
<link rel="preload" as="script" href="_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
4444

45-
<script src="_static/documentation_options.js?v=0c96e96c"></script>
45+
<script src="_static/documentation_options.js?v=9a529362"></script>
4646
<script src="_static/doctools.js?v=9a2dae69"></script>
4747
<script src="_static/sphinx_highlight.js?v=dc90522c"></script>
4848
<script src="_static/clipboard.min.js?v=a7894cd8"></script>
@@ -55,7 +55,7 @@
5555
<link rel="search" title="Search" href="search.html" />
5656
<meta name="viewport" content="width=device-width, initial-scale=1"/>
5757
<meta name="docsearch:language" content="en"/>
58-
<meta name="docsearch:version" content="0.4.3.post4" />
58+
<meta name="docsearch:version" content="0.4.4" />
5959
<meta name="docbuild:last-update" content="Mar 13, 2025"/>
6060
</head>
6161

‎_sources/backend/function_calling.ipynb

+122-122
Large diffs are not rendered by default.

‎_sources/backend/native_api.ipynb

+213-207
Large diffs are not rendered by default.

‎_sources/backend/offline_engine_api.ipynb

+440-443
Large diffs are not rendered by default.

‎_sources/backend/openai_api_completions.ipynb

+203-172
Large diffs are not rendered by default.

‎_sources/backend/openai_api_embeddings.ipynb

+68-62
Large diffs are not rendered by default.

‎_sources/backend/openai_api_vision.ipynb

+80-92
Large diffs are not rendered by default.

‎_sources/backend/send_request.ipynb

+109-91
Large diffs are not rendered by default.

‎_sources/backend/separate_reasoning.ipynb

+101-110
Large diffs are not rendered by default.

‎_sources/backend/speculative_decoding.ipynb

+151-169
Large diffs are not rendered by default.

‎_sources/backend/structured_outputs.ipynb

+116-122
Large diffs are not rendered by default.

‎_sources/developer/setup_github_runner.md

+2-2
Original file line numberDiff line numberDiff line change
@@ -11,9 +11,9 @@ docker pull nvidia/cuda:12.1.1-devel-ubuntu22.04
1111
# Nvidia
1212
docker run --shm-size 128g -it -v /tmp/huggingface:/hf_home --gpus all nvidia/cuda:12.1.1-devel-ubuntu22.04 /bin/bash
1313
# AMD
14-
docker run --rm --device=/dev/kfd --device=/dev/dri --group-add video --shm-size 128g -it -v /tmp/huggingface:/hf_home lmsysorg/sglang:v0.4.3.post4-rocm630 /bin/bash
14+
docker run --rm --device=/dev/kfd --device=/dev/dri --group-add video --shm-size 128g -it -v /tmp/huggingface:/hf_home lmsysorg/sglang:v0.4.4-rocm630 /bin/bash
1515
# AMD just the last 2 GPUs
16-
docker run --rm --device=/dev/kfd --device=/dev/dri/renderD176 --device=/dev/dri/renderD184 --group-add video --shm-size 128g -it -v /tmp/huggingface:/hf_home lmsysorg/sglang:v0.4.3.post4-rocm630 /bin/bash
16+
docker run --rm --device=/dev/kfd --device=/dev/dri/renderD176 --device=/dev/dri/renderD184 --group-add video --shm-size 128g -it -v /tmp/huggingface:/hf_home lmsysorg/sglang:v0.4.4-rocm630 /bin/bash
1717
```
1818

1919
### Step 2: Configure the runner by `config.sh`

‎_sources/frontend/frontend.ipynb

+212-209
Large diffs are not rendered by default.

‎_sources/start/install.md

+6-6
Original file line numberDiff line numberDiff line change
@@ -11,7 +11,7 @@ It is recommended to use uv to install the dependencies for faster installation:
1111
```bash
1212
pip install --upgrade pip
1313
pip install uv
14-
uv pip install "sglang[all]>=0.4.3.post4" --find-links https://flashinfer.ai/whl/cu124/torch2.5/flashinfer-python
14+
uv pip install "sglang[all]>=0.4.4" --find-links https://flashinfer.ai/whl/cu124/torch2.5/flashinfer-python
1515
```
1616

1717
**Quick Fixes to Common Problems**
@@ -27,7 +27,7 @@ uv pip install "sglang[all]>=0.4.3.post4" --find-links https://flashinfer.ai/whl
2727
## Method 2: From source
2828
```
2929
# Use the last release branch
30-
git clone -b v0.4.3.post4 https://github.com/sgl-project/sglang.git
30+
git clone -b v0.4.4 https://github.com/sgl-project/sglang.git
3131
cd sglang
3232
3333
pip install --upgrade pip
@@ -42,7 +42,7 @@ Note: For AMD ROCm system with Instinct/MI GPUs, do following instead:
4242

4343
```
4444
# Use the last release branch
45-
git clone -b v0.4.3.post4 https://github.com/sgl-project/sglang.git
45+
git clone -b v0.4.4 https://github.com/sgl-project/sglang.git
4646
cd sglang
4747
4848
pip install --upgrade pip
@@ -70,7 +70,7 @@ docker run --gpus all \
7070
Note: For AMD ROCm system with Instinct/MI GPUs, it is recommended to use `docker/Dockerfile.rocm` to build images, example and usage as below:
7171

7272
```bash
73-
docker build --build-arg SGL_BRANCH=v0.4.3.post4 -t v0.4.3.post4-rocm630 -f Dockerfile.rocm .
73+
docker build --build-arg SGL_BRANCH=v0.4.4 -t v0.4.4-rocm630 -f Dockerfile.rocm .
7474

7575
alias drun='docker run -it --rm --network=host --device=/dev/kfd --device=/dev/dri --ipc=host \
7676
--shm-size 16G --group-add video --cap-add=SYS_PTRACE --security-opt seccomp=unconfined \
@@ -79,11 +79,11 @@ alias drun='docker run -it --rm --network=host --device=/dev/kfd --device=/dev/d
7979
drun -p 30000:30000 \
8080
-v ~/.cache/huggingface:/root/.cache/huggingface \
8181
--env "HF_TOKEN=<secret>" \
82-
v0.4.3.post4-rocm630 \
82+
v0.4.4-rocm630 \
8383
python3 -m sglang.launch_server --model-path meta-llama/Llama-3.1-8B-Instruct --host 0.0.0.0 --port 30000
8484

8585
# Till flashinfer backend available, --attention-backend triton --sampling-backend pytorch are set by default
86-
drun v0.4.3.post4-rocm630 python3 -m sglang.bench_one_batch --batch-size 32 --input 1024 --output 128 --model amd/Meta-Llama-3.1-8B-Instruct-FP8-KV --tp 8 --quantization fp8
86+
drun v0.4.4-rocm630 python3 -m sglang.bench_one_batch --batch-size 32 --input 1024 --output 128 --model amd/Meta-Llama-3.1-8B-Instruct-FP8-KV --tp 8 --quantization fp8
8787
```
8888

8989
## Method 4: Using docker compose

‎_static/documentation_options.js

+1-1
Original file line numberDiff line numberDiff line change
@@ -1,5 +1,5 @@
11
const DOCUMENTATION_OPTIONS = {
2-
VERSION: '0.4.3.post4',
2+
VERSION: '0.4.4',
33
LANGUAGE: 'en',
44
COLLAPSE_INDEX: false,
55
BUILDER: 'html',

‎backend/custom_chat_template.html

+2-2
Original file line numberDiff line numberDiff line change
@@ -42,7 +42,7 @@
4242
<link rel="preload" as="script" href="../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
4343
<link rel="preload" as="script" href="../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
4444

45-
<script src="../_static/documentation_options.js?v=0c96e96c"></script>
45+
<script src="../_static/documentation_options.js?v=9a529362"></script>
4646
<script src="../_static/doctools.js?v=9a2dae69"></script>
4747
<script src="../_static/sphinx_highlight.js?v=dc90522c"></script>
4848
<script src="../_static/clipboard.min.js?v=a7894cd8"></script>
@@ -57,7 +57,7 @@
5757
<link rel="prev" title="Reasoning Parser" href="separate_reasoning.html" />
5858
<meta name="viewport" content="width=device-width, initial-scale=1"/>
5959
<meta name="docsearch:language" content="en"/>
60-
<meta name="docsearch:version" content="0.4.3.post4" />
60+
<meta name="docsearch:version" content="0.4.4" />
6161
<meta name="docbuild:last-update" content="Mar 13, 2025"/>
6262
</head>
6363

0 commit comments

Comments
 (0)