From Chunks to Blocks: Accelerating Uploads and Downloads on the Hub

Rearchitecting Hugging Face Uploads and Downloads

From DeepSpeed to FSDP and Back Again with Hugging Face Accelerate

Migrating the Hub from Git LFS to Xet

Hugging Face and AMD partner on accelerating state-of-the-art models for CPU and GPU platforms

From Scores to Gibbs Correctors: Accelerating Uniform-Rate Discrete Diffusion Models

Discrete diffusion models have achieved strong empirical performance in text and other symbolic domains, but, especially for uniform-rate models, they often require many steps to generate a single sample. Existing acceleration methods either rely on training additional quantities or suffer from slow mixing. In this...

Accelerating Document AI