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 supported59 Subscribers
View all SubscribersAdd a CodeTriage badge to vllm
Help out
- Issues
- [Model] Add native CTC word timestamps for Nemotron 3.5 Transcribe
- [Frontend][v0.30.0][Reference only] Backport Messages min_tokens and ignore_eos
- [Feature][HiSparse] Add backend hooks for private host allocation
- [RFC]: Decouple Physical PCP Topology from Per-Forward Attention Layout and Draft Execution Policy
- [Bug]: Resolve Triton kernel block sizes per platform instead of hardcoding them
- [ROCm][Quantization] Add FP8 block GEMM configs for gfx1100 (AMD_Radeon_Graphics)
- [Bugfix][Quantization] Honor the activation scheme in MXFP8 linear kernel selection
- [Perf][Attention] Cache FlexAttention K/V cache views per tensor
- [Perf][Mamba] Cache the conv_state transpose view per state-cache tensor
- [Rust Frontend] End reasoning at a tool-call opener like Python
- Docs
- Python not yet supported