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 supported51 Subscribers
View all SubscribersAdd a CodeTriage badge to vllm
Help out
- Issues
- [Model] Default quantized GLM KV cache to FP8 on NVIDIA
- [Bugfix][Kimi-K3] Convert developer role to system for updated K3 encoding rule
- [ROCm][DSv4] Enable aux-stream for c4a/c128a compressor overlap
- [ROCm][Quant] Accept pre-quantized activations in the AITER MXFP4 linear kernel
- [hw-agnostic] Implemented plumbing to support OOT plugins
- [ROCm][Quant][Kimi-K3] Fuse MXFP4 activation quant into Kimi-K3 attention epilogues
- [Bugfix][Multimodal] Guard FlashInfer empty queries for --mm-encoder-tp-mode data
- [glm] [pcp] fuse norm rope with pcp cache gather
- [Bug]: TurboQuant k8v4 + MTP speculative decoding silently produces degenerate output on hybrid GDN models (v0.27.1, stock)
- [Perf][Humming] Integrate fused SwiGLU FP8 quantization for W2
- Docs
- Python not yet supported