vllm
https://github.com/vllm-project/vllm
Python
A high-throughput and memory-efficient inference and serving engine for LLMs
Triage Issues!
When you volunteer to triage issues, you'll receive an email each day with a link to an open issue that needs help in this project. You'll also receive instructions on how to triage issues.
Triage Docs!
Receive a documented method or class from your favorite GitHub repos in your inbox every day. If you're really pro, receive undocumented methods or classes and supercharge your commit history.
Python not yet supported21 Subscribers
Add a CodeTriage badge to vllm
Help out
- Issues
- [Feature]: Optimize collectives in TP MoE case using torch.compile pass
- Support compressed-tensors W4A8 MoE checkpoints in GptOssModel weight loader for CPU
- [Bugfix]: Streaming i/o of batch files. Resolves #30268
- [Performance]: DeepSeek-V3.2 AWQ Performance is lower then i expected
- [Core] Add SLA-tiered scheduling (opt-in)
- feat: support video list inference
- [CI] Fix mypy for `vllm/lora`
- [Misc] Support amd-quark online fp8_ptpc quant
- [Bug]: NVFP4 Flashinfer CuteDSL MoE + DeepEP ll + VLLM_MOE_DP_CHUNK_SIZE=1024 + cudagraph numerical accuracy issue on B200
- fix: improve NIXL import safety and add UCX compatibility checks
- Docs
- Python not yet supported