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 supported13 Subscribers
Add a CodeTriage badge to vllm
Help out
- Issues
- [Feature]: Propose in docs a complete example of `pyproject.toml` to be used directly with `uv sync`
- [RFC]: Add a cache hit threshold to handle Preemptions in PD-Disaggregation and enable lightweight powerful P/D implementations
- [Installation]: 在amd rocm6.4.2设备上安装报错
- [Bug]: msgspec.DecodeError: MessagePack data is malformed: trailing characters (byte 198)
- [Misc][Test] Moe layer test
- [Bug]: torch.compile'd kv norm operation produces 4 kernels
- [WIP][Attention] Support attention sink used by gpt-oss in FlashInfer backend
- [Bug]: ValueError: Model architectures ['EagleQwen2ForCausalLM'] are not supported for now. vllm not loading with EAGLE spec decode using drafter model EAGLE-Qwen2-7B-Instruct and main model Qwen2-7B-Instruct
- [Bug]: vLLM fails to load k_scale and v_scale for FP8 quantized Qwen3-MOE models
- [Model] Funstep Step-Audio2 MM
- Docs
- Python not yet supported