AI BriefingHuggingFace Blog news
The latest AI and developer news from HuggingFace Blog, with short summaries.
Visit website →Hugging Face-IBM, watsonx.ai Collaboration
Hugging Face and IBM announced a collaboration for building enterprise AI through watsonx.ai.
Optimizing CPU LLM Inference with SmoothQuant
This covers how to use SmoothQuant technology to reduce LLM size and improve inference efficiency on Intel CPUs.
BigCode Reveals Large-Scale Data Deduplication Techniques
This explains large-scale near-duplicate deduplication techniques for improving the quality of LLM training data and their importance.
RWKV: A Combination of Transformer and RNN
RWKV, a new architecture that combines the efficiency of Transformer with the advantages of RNN, has been released.
Hugging Face Selected for CNIL Support
Hugging Face has been selected for the French data protection authority (CNIL)'s AI data protection support program.
Low-Latency Text Generation: 'Assisted Generation'
It introduces a new decoding technique that reduces text generation latency by up to 10x by solving the memory bandwidth problem, a bottleneck in LLM inference.
StarChat Released, Built on StarCoder
Introduces how to fine-tune StarCoder with ChatML and DeepSpeed to build the conversational coding assistant StarChat.
Hugging Face Releases Code LLM StarCoder
Hugging Face and ServiceNow collaborated to release StarCoder, a high-performance open-source code LLM.
TensorFlow-based TPU Training Guide Released
This guide walks through the end-to-end process of training a RoBERTa model from scratch using TensorFlow and TPU.
Databricks-HF launches Spark integration feature
Databricks and Hugging Face have launched a feature that directly converts Spark dataframes into Hugging Face datasets.
Hugging Face Launches Chinese-Language Blog
Hugging Face has launched a dedicated Chinese-language blog to collaborate with the Chinese AI community.
AWS Inferentia2 Optimization Support
Hugging Face now supports AWS Inferentia2 optimization, improving inference performance and cost efficiency for Transformers models.
Graph Classification with Transformers
Explains how to implement graph classification using the Transformers library and the Graphormer model.
Introducing Substra, a Federated Learning Framework
Substra is a federated learning framework that lets you train on distributed data without any data leakage.
Snorkel AI Strengthens Enterprise FM Utilization
Snorkel AI is expanding its enterprise foundation model development environment by leveraging Hugging Face's Inference Endpoint.
StackLLaMA: A Guide to Training LLaMA with RLHF
Hugging Face released the StackLLaMA model and a guide for training LLaMA using RLHF.
Hugging Face Introduces Ethical ML Tags
Hugging Face is introducing new ethics tags and safeguards to support the ethical use of AI models.
Accelerating Stable Diffusion Inference on Intel CPUs
This introduces a method for speeding up Stable Diffusion inference by more than 2x on Intel Sapphire Rapids CPUs using OpenVINO.
Accelerating BLOOMZ Inference on Gaudi2
Introduces how to efficiently deploy and run inference on the 176B-scale BLOOMZ model using 🤗 Optimum Habana on Habana Gaudi2 accelerators.
Implementing Federated Learning with Hugging Face and Flower
A technical tutorial on implementing federated learning for language models by combining Hugging Face with the Flower framework.