io-aware

Flash-dLLM is a training-free inference acceleration framework for diffusion LLMs that uses IO-aware KV caching and parallel decoding to achieve significant speedups and memory efficiency improvements.

Flash-KMeans is an IO-aware implementation of exact KMeans that redesigns the algorithm around modern GPU bottlenecks, achieving 33x speedup over cuML and 200x over FAISS by eliminating redundant memory reads and writes.

TIDE is a lossless inference system for diffusion large language models that leverages temporal stability of expert activations to reduce I/O overhead and computation, achieving up to 1.4-1.5x throughput improvements on single GPU-CPU systems.