AI BriefingHuggingFace Blog news
The latest AI and developer news from HuggingFace Blog, with short summaries.
Visit website →Diffusers Adds LoRA Training Support for Stable Diffusion
It introduces a method of applying LoRA technology to Stable Diffusion to speed up training and dramatically reduce model size.
Optimum-ONNX Runtime Boosts Training Speed
By integrating Hugging Face's Optimum library with ONNX Runtime, training speed for transformer models has been improved by up to 130%.
Conversational AI Agent Technology and Model Comparison
This analyzes and compares the technical characteristics and training methodologies of major conversational AI models, including ChatGPT.
Universal Segmentation Based on Mask2Former
Introducing a universal model that performs instance, semantic, and panoptic segmentation in an integrated way through Mask2Former and OneFormer.
Hugging Face begins supporting PaddlePaddle models
The Hugging Face Hub now supports models from PaddlePaddle, Baidu's open-source deep learning framework.
Intel Sapphire Rapids Accelerates PyTorch
Introduces how to accelerate PyTorch Transformer training using the AMX technology of Intel Sapphire Rapids.
CLIPSeg: Zero-shot Segmentation
This introduces how CLIPSeg, a CLIP-based zero-shot image segmentation model, works and how to use it.
Hugging Face Launches Model Card Creation Tool
Hugging Face has released a model card creator tool and guide for model documentation standards.
Hugging Face Audio Dataset Guide
This guide explains how to efficiently explore, load, and process audio datasets using the Hugging Face `datasets` library.
Hugging Face, Guide to Addressing ML Bias
Introduces Hugging Face's tools and guidance for identifying and mitigating bias throughout the entire machine learning development process.
Gaudi2 delivers 2x faster performance than Nvidia A100
Habana Gaudi2 recorded roughly 2x faster performance than Nvidia A100 80GB in BERT training and Stable Diffusion inference, among others.
Bumblebee Library Released for Elixir
The Bumblebee library has been released, enabling Hugging Face models such as GPT2 and Stable Diffusion to run in the Elixir environment.
Key Stages and Mechanics of RLHF
This explains the concept of RLHF, which optimizes language models through human feedback, and its three-stage training process.
Guide to Protein Language Models and Using ESMFold
This covers how to understand and train on protein sequence data from the perspective of language models, and how to predict protein structures using ESMFold.
Transformer-Based Probabilistic Time Series Forecasting Methodology
It explains a method for performing probabilistic time series forecasting that incorporates uncertainty by leveraging the Encoder-Decoder structure of the Transformer.
SD Support for Apple Silicon
Reveals how to run Stable Diffusion on Apple Silicon via Core ML and releases the converted models.
VQ-Diffusion: A Discrete Latent Space Model
This covers the technical features and implementation of VQ-Diffusion, a conditional latent diffusion model that operates in a discrete latent space.
Hugging Face Inference Solutions Guide
This introduces the step-by-step model inference and deployment solutions offered by Hugging Face.
Document AI Key Tasks and Model Guide
A technical guide summarizing the key tasks of Document AI for document understanding and how to leverage multimodal models.
Privacy-Preserving Sentiment Analysis Based on Homomorphic Encryption
Introduces how to perform Transformer-based sentiment analysis on encrypted data using Concrete-ML.