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 supported62 Subscribers
View all SubscribersAdd a CodeTriage badge to vllm
Help out
- Issues
- [Bug]: pythonic tool parsers stream wrong numbers in tool call arguments
- [RFC] [KV Offload]: Store triggers and store modes for the offloading connector (lazy offloading)
- [RFC]: vLLM Simulator: CPU-Based CI and Hardware-Aware Performance Simulation
- [CI][Feature] Meta-device mode for CPU-only end-to-end engine CI
- [Bugfix] Preserve integer tool argument precision
- [Model] Share the Inkling Triton implementation via inkling/common/
- [Kernel] Coalesce reshape_and_cache stores by taking a KV block at a time
- [Multimodal] Add GLM-5.3-Flash and DeepSeek V4/V4.1 routing-metadata adapters
- [Bug]: Chat Completions returns `annotations: null` (typed as one Annotation, not a list) and `usage.prompt_tokens_details: null`; strict OpenAI-schema validators reject both
- [ROCm][Perf] Fuse GDN decode (conv1d + delta rule + gated norm) into AITER's gluon kernel (Qwen3.8-27B)
- Docs
- Python not yet supported