Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

[V1] Avoid list creation in input preparation #12457

Merged
merged 2 commits into from
Jan 27, 2025
Merged

[V1] Avoid list creation in input preparation #12457

merged 2 commits into from
Jan 27, 2025

Conversation

WoosukKwon
Copy link
Collaborator

A Numpy trick to avoid Python list operations. I think this is especially useful when the batch has many (~1K) decode requests.

@github-actions GitHub Actions
Copy link

👋 Hi! Thank you for contributing to the vLLM project.
Just a reminder: PRs would not trigger full CI run by default. Instead, it would only run fastcheck CI which starts running only a small and essential subset of CI tests to quickly catch errors. You can run other CI tests on top of those by going to your fastcheck build on Buildkite UI (linked in the PR checks section) and unblock them. If you do not have permission to unblock, ping simon-mo or khluu to add you in our Buildkite org.

Once the PR is approved and ready to go, your PR reviewer(s) can run CI to test the changes comprehensively before merging.

To run CI, PR reviewers can do one of these:

  • Add ready label to the PR
  • Enable auto-merge.

🚀

@WoosukKwon
Fix
31b1a8a 
Signed-off-by: Woosuk Kwon <[email protected]>
@WoosukKwon WoosukKwon added the ready ONLY add when PR is ready to merge/full CI is needed label Jan 27, 2025
@WoosukKwon WoosukKwon merged commit 28e0750 into main Jan 27, 2025
57 of 59 checks passed
@WoosukKwon WoosukKwon deleted the v1-opt-np-prep branch January 27, 2025 03:57
tjtanaa added a commit to EmbeddedLLM/vllm that referenced this pull request Jan 27, 2025
@tjtanaa @ywang96
@youkaichao @MohitIntel @jsato8094 @LucasWilkinson @DarkLight1337 @ElizaWszola @divakar-amd @Isotr0py @lsy323 @youngkent @mhendrey @ShangmingCai @hmellor @terrytangyuan @heheda12345 @wangxiyuan @tlrmchlsmth @mgoin @K-Mistele @WoosukKwon @pooyadavoodi
[MFM] Merge changes from vllm-project/vllm main (#28)
e8e548c 
* [Misc] Use VisionArena Dataset for VLM Benchmarking (vllm-project#12389)

Signed-off-by: Roger Wang <[email protected]>

* [ci/build] fix wheel size check (vllm-project#12396)

Signed-off-by: youkaichao <[email protected]>

* [Hardware][Gaudi][Doc] Add missing step in setup instructions (vllm-project#12382)

* [ci/build] sync default value for wheel size (vllm-project#12398)

Signed-off-by: youkaichao <[email protected]>

* [Misc] Enable proxy support in benchmark script (vllm-project#12356)

Signed-off-by: Junichi Sato <[email protected]>

* [Bugfix][Kernel] Fix CUDA 11.8 being broken by FA3 build (vllm-project#12375)

Signed-off-by: Lucas Wilkinson <[email protected]>

* [Misc] Remove deprecated code (vllm-project#12383)

Signed-off-by: DarkLight1337 <[email protected]>

* [Bugfix][Kernel] FA3 Fix - RuntimeError: This flash attention build only supports pack_gqa (for build size reasons). (vllm-project#12405)

Signed-off-by: Lucas Wilkinson <[email protected]>

* [Bugfix][Kernel] Fix moe align block issue for mixtral (vllm-project#12413)

* [Bugfix] Fix BLIP-2 processing (vllm-project#12412)

Signed-off-by: DarkLight1337 <[email protected]>

* [ROCm][MoE] MI300 tuned configs Mixtral-8x(7B,22B) | fp16, fp8 (vllm-project#12408)

Signed-off-by: Divakar Verma <[email protected]>

* [Misc] Add FA2 support to ViT MHA layer (vllm-project#12355)

Signed-off-by: Isotr0py <[email protected]>

* [TPU][CI] Update torchxla version in requirement-tpu.txt (vllm-project#12422)

Signed-off-by: Siyuan Liu <[email protected]>

* [Misc][Bugfix] FA3 support to ViT MHA layer (vllm-project#12435)

Signed-off-by: Roger Wang <[email protected]>
Signed-off-by: Isotr0py <[email protected]>
Co-authored-by: Isotr0py <[email protected]>

* [V1][Perf] Reduce scheduling overhead in model runner after cuda sync (vllm-project#12094)

Signed-off-by: Keyun Tong <[email protected]>

* [V1][Bugfix] Fix assertion when mm hashing is turned off (vllm-project#12439)

Signed-off-by: Roger Wang <[email protected]>

* [Misc] Revert FA on ViT vllm-project#12355 and vllm-project#12435 (vllm-project#12445)

* [Frontend] generation_config.json for  maximum tokens(vllm-project#12242)

Signed-off-by: Matthew Hendrey <[email protected]>
Signed-off-by: Shangming Cai <[email protected]>
Signed-off-by: youkaichao <[email protected]>
Signed-off-by: Harry Mellor <[email protected]>
Signed-off-by: Yuan Tang <[email protected]>
Signed-off-by: Isotr0py <[email protected]>
Signed-off-by: DarkLight1337 <[email protected]>
Signed-off-by: Chen Zhang <[email protected]>
Signed-off-by: wangxiyuan <[email protected]>
Co-authored-by: shangmingc <[email protected]>
Co-authored-by: youkaichao <[email protected]>
Co-authored-by: Harry Mellor <[email protected]>
Co-authored-by: Yuan Tang <[email protected]>
Co-authored-by: Isotr0py <[email protected]>
Co-authored-by: Cyrus Leung <[email protected]>
Co-authored-by: Chen Zhang <[email protected]>
Co-authored-by: wangxiyuan <[email protected]>

* [Bugfix] Disable w16a16 2of4 sparse CompressedTensors24 (vllm-project#12417)

Signed-off-by: Tyler Michael Smith <[email protected]>
Co-authored-by: mgoin <[email protected]>

* [Bugfix/CI] Fix broken kernels/test_mha.py (vllm-project#12450)

* [Bugfix][Kernel] Fix perf regression caused by PR vllm-project#12405 (vllm-project#12434)

Signed-off-by: Lucas Wilkinson <[email protected]>

* [Build/CI] Fix libcuda.so linkage (vllm-project#12424)

Signed-off-by: Tyler Michael Smith <[email protected]>

* [Frontend] Rerank API (Jina- and Cohere-compatible API)  (vllm-project#12376)

Signed-off-by: Kyle Mistele <[email protected]>

* [DOC] Add link to vLLM blog (vllm-project#12460)

Signed-off-by: Yuan Tang <[email protected]>

* [V1] Avoid list creation in input preparation (vllm-project#12457)

Signed-off-by: Woosuk Kwon <[email protected]>

* [Frontend] Support scores endpoint in run_batch (vllm-project#12430)

Signed-off-by: Pooya Davoodi <[email protected]>

* [Bugfix] Fix Granite 3.0 MoE model loading (vllm-project#12446)

Signed-off-by: DarkLight1337 <[email protected]>

* [Bugfix] Fix missing seq_start_loc in xformers prefill metadata (vllm-project#12464)

Signed-off-by: Isotr0py <[email protected]>

* [V1][Minor] Minor optimizations for update_from_output (vllm-project#12454)

Signed-off-by: Woosuk Kwon <[email protected]>

* [Bugfix] Fix gpt2 GGUF inference (vllm-project#12467)

Signed-off-by: Isotr0py <[email protected]>

---------

Signed-off-by: Roger Wang <[email protected]>
Signed-off-by: youkaichao <[email protected]>
Signed-off-by: Junichi Sato <[email protected]>
Signed-off-by: Lucas Wilkinson <[email protected]>
Signed-off-by: DarkLight1337 <[email protected]>
Signed-off-by: Divakar Verma <[email protected]>
Signed-off-by: Isotr0py <[email protected]>
Signed-off-by: Siyuan Liu <[email protected]>
Signed-off-by: Keyun Tong <[email protected]>
Signed-off-by: Matthew Hendrey <[email protected]>
Signed-off-by: Shangming Cai <[email protected]>
Signed-off-by: Harry Mellor <[email protected]>
Signed-off-by: Yuan Tang <[email protected]>
Signed-off-by: Chen Zhang <[email protected]>
Signed-off-by: wangxiyuan <[email protected]>
Signed-off-by: Tyler Michael Smith <[email protected]>
Signed-off-by: Kyle Mistele <[email protected]>
Signed-off-by: Woosuk Kwon <[email protected]>
Signed-off-by: Pooya Davoodi <[email protected]>
Co-authored-by: Roger Wang <[email protected]>
Co-authored-by: youkaichao <[email protected]>
Co-authored-by: Mohit Deopujari <[email protected]>
Co-authored-by: Junichi Sato <[email protected]>
Co-authored-by: Lucas Wilkinson <[email protected]>
Co-authored-by: Cyrus Leung <[email protected]>
Co-authored-by: ElizaWszola <[email protected]>
Co-authored-by: Divakar Verma <[email protected]>
Co-authored-by: Isotr0py <[email protected]>
Co-authored-by: Siyuan Liu <[email protected]>
Co-authored-by: Isotr0py <[email protected]>
Co-authored-by: Keyun Tong <[email protected]>
Co-authored-by: Matthew Hendrey <[email protected]>
Co-authored-by: shangmingc <[email protected]>
Co-authored-by: Harry Mellor <[email protected]>
Co-authored-by: Yuan Tang <[email protected]>
Co-authored-by: Chen Zhang <[email protected]>
Co-authored-by: wangxiyuan <[email protected]>
Co-authored-by: Tyler Michael Smith <[email protected]>
Co-authored-by: mgoin <[email protected]>
Co-authored-by: Kyle Mistele <[email protected]>
Co-authored-by: Woosuk Kwon <[email protected]>
Co-authored-by: Pooya Davoodi <[email protected]>
tjtanaa pushed a commit to EmbeddedLLM/vllm that referenced this pull request Jan 28, 2025
@WoosukKwon @tjtanaa
[V1] Avoid list creation in input preparation (vllm-project#12457)
7b52511 
Signed-off-by: Woosuk Kwon <[email protected]>
rasmith pushed a commit to rasmith/vllm that referenced this pull request Jan 30, 2025
@WoosukKwon @rasmith
[V1] Avoid list creation in input preparation (vllm-project#12457)
712ab9e 
Signed-off-by: Woosuk Kwon <[email protected]>
Isotr0py pushed a commit to Isotr0py/vllm that referenced this pull request Feb 2, 2025
@WoosukKwon @Isotr0py
[V1] Avoid list creation in input preparation (vllm-project#12457)
078126f 
Signed-off-by: Woosuk Kwon <[email protected]>
Signed-off-by: Isotr0py <[email protected]>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@robertgshaw2-redhat robertgshaw2-redhat Awaiting requested review from robertgshaw2-redhat robertgshaw2-redhat is a code owner

@njhill njhill Awaiting requested review from njhill njhill is a code owner

@ywang96 ywang96 Awaiting requested review from ywang96 ywang96 is a code owner

@comaniac comaniac Awaiting requested review from comaniac comaniac is a code owner

@alexm-redhat alexm-redhat Awaiting requested review from alexm-redhat alexm-redhat is a code owner

Assignees
No one assigned
Labels
ready ONLY add when PR is ready to merge/full CI is needed
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
1 participant
@WoosukKwon