vLLM Windows CUDA support [tested] #2158
Open
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
change vllm installed check by transformers utils function
change vllm installed check by transformers utils function
|
Tested on windows 11, working as intended. |
|
Oh ok! Great vLLM works on Windows - it's maybe best we also add a print statement showing you can use https://github.com/SystemPanic/vllm-windows! Maybe we should add it in the readme! |
Datta0
reviewed
Apr 15, 2025
unsloth/models/loader.py
Outdated
Comment on lines
342
to
345
| from transformers.utils.import_utils import _is_package_available | ||
| _vllm_available = _is_package_available("vllm") | ||
| if _vllm_available == False: | ||
| print("Unsloth: vLLM is not installed! Will use Unsloth inference!") |
There was a problem hiding this comment.
Minor suggestion: can we have this check at the top of loader.py (like where we check transformers versions for model support)
And maybe set it as a constant that we can later reuse in llama.py as well instead of duplicating it?
There was a problem hiding this comment.
now a new function is_vLLM_available was added to utils.
Datta0
reviewed
Apr 16, 2025
| from transformers.utils.import_utils import _is_package_available | ||
| _vllm_available = _is_package_available("vllm") | ||
| if _vllm_available == False: | ||
| if is_vLLM_available() == False: |
There was a problem hiding this comment.
NIT: General python way is to use if not is_vllm_available()
But this is fine as well...
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
vLLM Windows CUDA support
vllm-windows Windows wheels by SystemPanic
install
conda create -n vllm python=3.12 conda activate vllm pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu124 pip install https://github.com/SystemPanic/vllm-windows/releases/download/v0.8.1/vllm-0.8.1+cu124-cp312-cp312-win_amd64.whl pip install https://github.com/SystemPanic/flashinfer-windows/releases/download/v0.2.3/flashinfer_python-0.2.3+cu124torch2.6-cp312-cp312-win_amd64.whl pip install --upgrade pillow pip install --upgrade pandas pip install --upgrade triton-windows pip install grpcio==1.71.0 pip install "unsloth[windows] @ git+https://github.com/fenglui/unsloth.git" pip install --no-deps git+https://github.com/huggingface/transformers.git pip install trl==0.15.2training test
download https://github.com/unslothai/notebooks/blob/main/nb/Qwen2.5_(3B)-GRPO.ipynb
remove the installation code block
add code block at top
then you can run the rest of the training code.
if the vllm serve faild, add env var "USE_LIBUV" and set value "0" to your windows system var
and that's it, we can run unsloth training with vllm support with Windows.