[MISC] Upgrade dependency to PyTorch 2.3.1

[comaniac]

PyTorch 2.3.1 just released, and this PR upgrades the dependency to 2.3.1.
The most important reason of this upgrade is PyTorch strictly depends on a particular triton version, and PyTorch 2.3.0 depends on triton 2.3.0. However, @pcmoritz pointed out a performance bug in triton 2.3.0 and it has been fixed in triton 2.3.1, so in order to achieve the best Mixtral FP8 performance, we have to use triton 2.3.1, which results in version conflict if we depend on PyTorch 2.3.0.

cc @Yard1 @pcmoritz @robertgshaw2-neuralmagic @simon-mo

FIX #4509
FIX #5535
FIX #5579
FIX #5705

BEFORE SUBMITTING, PLEASE READ THE CHECKLIST BELOW AND FILL IN THE DESCRIPTION ABOVE

---

PR Checklist (Click to Expand)

Thank you for your contribution to vLLM! Before submitting the pull request, please ensure the PR meets the following criteria. This helps vLLM maintain the code quality and improve the efficiency of the review process.

PR Title and Classification

Only specific types of PRs will be reviewed. The PR title is prefixed appropriately to indicate the type of change. Please use one of the following:

• [Bugfix] for bug fixes.
• [CI/Build] for build or continuous integration improvements.
• [Doc] for documentation fixes and improvements.
• [Model] for adding a new model or improving an existing model. Model name should appear in the title.
• [Frontend] For changes on the vLLM frontend (e.g., OpenAI API server, LLM class, etc.)
• [Kernel] for changes affecting CUDA kernels or other compute kernels.
• [Core] for changes in the core vLLM logic (e.g., LLMEngine, AsyncLLMEngine, Scheduler, etc.)
• [Hardware][Vendor] for hardware-specific changes. Vendor name should appear in the prefix (e.g., [Hardware][AMD]).
• [Misc] for PRs that do not fit the above categories. Please use this sparingly.

Note: If the PR spans more than one category, please include all relevant prefixes.

Code Quality

The PR need to meet the following code quality standards:

• We adhere to Google Python style guide and Google C++ style guide.
• Pass all linter checks. Please use format.sh to format your code.
• The code need to be well-documented to ensure future contributors can easily understand the code.
• Include sufficient tests to ensure the project to stay correct and robust. This includes both unit tests and integration tests.
• Please add documentation to docs/source/ if the PR modifies the user-facing behaviors of vLLM. It helps vLLM user understand and utilize the new features or changes.

Notes for Large Changes

Please keep the changes as concise as possible. For major architectural changes (>500 LOC excluding kernel/data/config/test), we would expect a GitHub issue (RFC) discussing the technical design and justification. Otherwise, we will tag it with rfc-required and might not go through the PR.

What to Expect for the Reviews

The goal of the vLLM team is to be a transparent reviewing machine. We would like to make the review process transparent and efficient and make sure no contributor feel confused or frustrated. However, the vLLM team is small, so we need to prioritize some PRs over others. Here is what you can expect from the review process:

• After the PR is submitted, the PR will be assigned to a reviewer. Every reviewer will pick up the PRs based on their expertise and availability.
• After the PR is assigned, the reviewer will provide status update every 2-3 days. If the PR is not reviewed within 7 days, please feel free to ping the reviewer or the vLLM team.
• After the review, the reviewer will put an action-required label on the PR if there are changes required. The contributor should address the comments and ping the reviewer to re-review the PR.
• Please respond to all comments within a reasonable time frame. If a comment isn't clear or you disagree with a suggestion, feel free to ask for clarification or discuss the suggestion.

Thank You

Finally, thank you for taking the time to read these guidelines and for your interest in contributing to vLLM. Your contributions make vLLM a great tool for everyone!

[robertgshaw2-neuralmagic]

I think you also need requirements-cuda.txt: https://github.com/vllm-project/vllm/blob/main/requirements-cuda.txt#L7

Do we need to update vllm-flash-attention as well, or are we okay with minor, minor version?
I think that xformers has not released a new package yet (latest stable is xformers==0.0.26.post1, which uses 2.3.0)

[comaniac]

Thanks for pointing out. I just changed all required places. Meanwhile, yeah we do need xFormers and vllm-flash-attn...

[WoosukKwon]

@comaniac vllm-flash-attn v2.5.9.post1 was built for PyTorch v2.3.1 and is now available in PyPI: https://pypi.org/project/vllm-flash-attn/

[comaniac]

Pending xformers to release a version against torch 2.3.1. Tracking issue facebookresearch/xformers#1052

Closing #4509

[comaniac]

CI passed, but need to double check manually whether FlashInfer supports torch 2.3.1.

[comaniac]

• Checked that FlashInfer works with PyTorch 2.3.1.
• Performance benchmark (https://buildkite.com/vllm/performance-benchmark/builds/4569#) shows no regression compared to https://simon-mo-workspace.observablehq.cloud/vllm-dashboard-v0/perf

This PR should be good to go.

cc @Yard1 @robertgshaw2-neuralmagic @WoosukKwon @simon-mo @DarkLight1337