vllm

Форк

Описание

A high-throughput and memory-efficient inference and serving engine for LLMs

amd

1.05 GiB

Языки

Python88,1%
Cuda6,4%
C++3,8%
Shell1%
CMake0,3%
C0,3%
Остальные0,1%

wang.yuqi
Revert "[Startup] Parallelize torch/transformers import + weight prefetch + forkserver prewarm" (#40438)
21 апр 2026, 11:47
Не верифицирован
21 апр 2026, 11:473975eb6

.buildkite

[CI][EPLB] Add Async EPLB end-to-end integration test to CI (#40168 )

2 месяца назад

.gemini

Configure Gemini (#20971 )

год назад

.github

Add @bbrowning to CODEOWNERS (#40141 )

2 месяца назад

benchmarks

[vLLM IR] Add IR op testing and benchmarking infrastructure (#40167 )

2 месяца назад

cmake

[CPU][RISC-V] Support multiple RVV VLEN targets via compile-time dispatch (#39478 )

2 месяца назад

csrc

[CPU][RISC-V] Support multiple RVV VLEN targets via compile-time dispatch (#39478 )

2 месяца назад

docker

Enable building MoRI with AMD AINIC stack (#38371 )

2 месяца назад

docs

[Deprecation] Deprecate cprofile and cprofile_context (#39100 )

2 месяца назад

examples

[MM][Misc] Support image+video mixed inputs (per prompt) for VLM examples (#40335 )

2 месяца назад

requirements

Update flashinfer to 0.6.8 (#39959 )

2 месяца назад

scripts

[Kernel] [Helion] [12/N] Use FakeTensorMode to avoid GPU allocation during config key computation (#36563 )

3 месяца назад

tests

[Bugfix] Normalize malformed dict prompts that carry token IDs in `prompt` (#40339 )

2 месяца назад

tools

Add structure to `requirements/` directory (#39024 )

2 месяца назад

vllm

Revert "[Startup] Parallelize torch/transformers import + weight prefetch + forkserver prewarm" (#40438 )

2 месяца назад

.clang-format

[CI/Build] Enforce style for C++ and CUDA code with `clang-format` (#4722 )

2 года назад

.coveragerc

Update coveragerc and add codecov.yml for path fixes (#26435 )

8 месяцев назад

.dockerignore

[CI/Build] remove .github from .dockerignore, add dirty repo check (#9375 )

2 года назад

.git-blame-ignore-revs

Ignore large reformatting PRs in `git blame` (#26690 )

8 месяцев назад

.gitignore

Add structure to `requirements/` directory (#39024 )

2 месяца назад

.markdownlint.yaml

[Docs] Enable some more markdown lint rules for the docs (#28731 )

7 месяцев назад

.pre-commit-config.yaml

Add structure to `requirements/` directory (#39024 )

2 месяца назад

.readthedocs.yaml

Disable docs build skipping until a better solution is found (#36790 )

3 месяца назад

.shellcheckrc

[CI/Build] Add shell script linting using shellcheck (#7925 )

2 года назад

.yapfignore

Add the support for the qwen3 next model (a hybrid attention model). (#24526 )

9 месяцев назад

AGENTS.md

Add structure to `requirements/` directory (#39024 )

2 месяца назад

CLAUDE.md

Add `AGENTS.md` (#36877 )

3 месяца назад

CMakeLists.txt

[Kernel] Add MXFP4 W4A4 CUTLASS MoE kernel for SM100 (#37463 )

2 месяца назад

CODE_OF_CONDUCT.md

[CI/Build] Auto-fix Markdown files (#12941 )

год назад

CONTRIBUTING.md

[Doc] Reorganize user guide (#18661 )

год назад

DCO

[Doc] Add the DCO to CONTRIBUTING.md (#9803 )

2 года назад

LICENSE

Add Apache-2.0 license (#102 )

3 года назад

MANIFEST.in

[V0 deprecation] Deprecate V0 Neuron backend (#21159 )

9 месяцев назад

README.md

[Docs] Update README (#39251 )

2 месяца назад

RELEASE.md

[Doc] Update release docs (#31799 )

5 месяцев назад

SECURITY.md

Enhance the pre-notification policy (#23532 )

9 месяцев назад

codecov.yml

Update coveragerc and add codecov.yml for path fixes (#26435 )

8 месяцев назад

mkdocs.yaml

Automatically add links to API docs for matching strings in docs (#37434 )

2 месяца назад

pyproject.toml

[FEAT] [Perf] [Gemma4] Fused Gemma4 Routing Function Triton (#39083 )

2 месяца назад

setup.py

[ZenCPU] AMD Zen CPU Backend with supported dtypes via zentorch weekly (#39967 )

2 месяца назад

use_existing_torch.py

Bugfix: `use_existing_torch.py`: Glob recursive subdirs in requirements (fixes #39024 ) (#39793 )

2 месяца назад

README.md

vllm

Описание

Языки

wang.yuqiRevert "[Startup] Parallelize torch/transformers import + weight prefetch + forkserver prewarm" (#40438)21 апр 2026, 11:47Не верифицирован21 апр 2026, 11:473975eb6

wang.yuqi
Revert "[Startup] Parallelize torch/transformers import + weight prefetch + forkserver prewarm" (#40438)
21 апр 2026, 11:47
Не верифицирован
21 апр 2026, 11:473975eb6