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 supported24 Subscribers
Add a CodeTriage badge to vllm
Help out
- Issues
- [Model] Support per-layer sliding window attention for Qwen3
- [Core] Use standalone autograd_cache_key for compilation dedup optimization
- [Kernel] Batch invariant NVFP4 Triton MoE backend
- [Bugfix][V1] Retain encoder cache entries for multimodal models under async scheduling
- [Bugfix] Make completions response construction use normalized per-prompt max_tokens
- [Fix] Sync gemma4 chat template from hf
- [Kernel][UX] Add `--linear-backend` arg for linear kernel selection
- [Feature]: SubSpec — Lossless Training-Free Speculative Decoding for CPU-Offloaded LLMs via Quantized Substitute Draft
- [Installation]: Vllm doesnt have any failed installation handling.
- [Bug]: vllm 0.19.0, gemma4, The format of the tool call returned by vllm is incorrect.
- Docs
- Python not yet supported