vllm: [Installation]: Could not find a version that satisfies the requirement xgrammar>=0.1.6; platform_machine == "x86_64" (from vllm) (from versions: none)

Your current environment

 # 创建环境
 conda create -n online_model_v4  python=3.10.13
 
 # 激活环境
 conda activate online_model_v4


pip install vllm==0.6.6.post1


How you are installing vllm

pip install vllm==0.6.6.post1




问题
Collecting prometheus-fastapi-instrumentator>=7.0.0 (from vllm==0.6.6.post1)
  Using cached https://pypi.tuna.tsinghua.edu.cn/packages/59/66/2e93a8f56adb51ede41d0ef5f4f0277522acc4adc87937f5457b7b5692a8/prometheus_fastapi_instrumentator-7.0.0-py3-none-any.whl (19 kB)
Collecting tiktoken>=0.6.0 (from vllm==0.6.6.post1)
  Using cached https://pypi.tuna.tsinghua.edu.cn/packages/2e/28/cf3633018cbcc6deb7805b700ccd6085c9a5a7f72b38974ee0bffd56d311/tiktoken-0.8.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (1.2 MB)
Collecting lm-format-enforcer<0.11,>=0.10.9 (from vllm==0.6.6.post1)
  Using cached https://pypi.tuna.tsinghua.edu.cn/packages/c1/01/e78fdf09de2b4e7750a402eaa4f6783c7215ededd4bc6fe4a3f6d69c49da/lm_format_enforcer-0.10.9-py3-none-any.whl (43 kB)
Collecting outlines==0.1.11 (from vllm==0.6.6.post1)
  Using cached https://pypi.tuna.tsinghua.edu.cn/packages/13/b4/99ea4a122bef60e3fd6402d19665aff1f928e0daf8fac3044d0b73f72003/outlines-0.1.11-py3-none-any.whl (87 kB)
Collecting lark==1.2.2 (from vllm==0.6.6.post1)
  Using cached https://pypi.tuna.tsinghua.edu.cn/packages/2d/00/d90b10b962b4277f5e64a78b6609968859ff86889f5b898c1a778c06ec00/lark-1.2.2-py3-none-any.whl (111 kB)
INFO: pip is looking at multiple versions of vllm to determine which version is compatible with other requirements. This could take a while.
ERROR: Could not find a version that satisfies the requirement xgrammar>=0.1.6; platform_machine == "x86_64" (from vllm) (from versions: none)
ERROR: No matching distribution found for xgrammar>=0.1.6; platform_machine == "x86_64"

Before submitting a new issue…

  • Make sure you already searched for relevant issues, and asked the chatbot living at the bottom right corner of the documentation page, which can answer lots of frequently asked questions.

About this issue

  • Original URL
  • State: closed
  • Created 2 years ago
  • Reactions: 7
  • Comments: 38 (10 by maintainers)

Commits related to this issue

Most upvoted comments

I also solved the problem by installing xgrammar from the source code. The main thing to note is that the versions of cmake and gcc need to be high enough. I used cmake-3.23 and gcc-12.1 respectively.

pip install pybind11
git clone --recursive https://github.com/mlc-ai/xgrammar.git
mkdir build && cd build/
cmake ..
make -j8
cd ../python/
pip install -e .

I’m in a similar situation. Installing xgrammar from source worked for me.

# based on https://xgrammar.mlc.ai/docs/start/install

<install git lfs>

python -m pip venv venv
source venv/bin/activate

<install appropriate pytorch version, yours may be different>
python -m pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu124

python -m pip install ninja pybind11
python -m pip install pre-commit

git clone --recursive https://github.com/mlc-ai/xgrammar.git && cd xgrammar
pre-commit install

mkdir build && cd build
cmake .. -G Ninja
ninja

cd ../python
python -m pip install .

# optionally, add to path
# echo "export PATH=\"$(pwd):\$PATH\"" >> ~/.bashrc

@russellb I think it is. My cluster server got a lower version of glibc (<2). All the wheel for xgrammar here are built with glibc>2.27. I downloaded the wheel with wget and pip install the wheel, ERROR: xgrammar-0.1.9-cp310-cp310-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl is not a supported wheel on this platform. The platform is only mismatched on glibc version. I patched my interpreter with patchelf on glibc 2.27 (because sudo/root is not available) and then just pip install xgrammar, it worked. Hence, the installation error seems not from the f**king internet ‘‘wall’’ of china, but a mismatched version of glibc.

This should be resolved as of #14590.

The next release, v0.8.0, will include the fix.

Can those facing this issue try xgrammar 0.1.14 to see if that resolves your issue?

have you tried building xgrammar yourself instead of installing a wheel to see if you can build a version compatible with your environment?