Paged Attention Python Packages

grouped-nf4-gemm

4-bit GPU kernels that compute directly on packed Mixture-of-Experts weights, for training and inference.

wllm-server

A production-grade, vLLM-style LLM inference server built natively for Windows. PagedAttention, continuous batching, CUDA graphs, and an OpenAI-compatible API, no WSL2 required.

langchain-pced

LangChain integration for Parallel Context-of-Experts Decoding (PCED)

grouped-mxfp4-gemm

4-bit GPU kernels that compute directly on packed Mixture-of-Experts weights, for training and inference.

gnf4

4-bit GPU kernels that compute directly on packed Mixture-of-Experts weights, for training and inference.

nf4gemm

4-bit GPU kernels that compute directly on packed Mixture-of-Experts weights, for training and inference.