AI BriefingHuggingFace Blog news
The latest AI and developer news from HuggingFace Blog, with short summaries.
Visit website →Hugging Face Unveils Quantization Backend 'Quanto'
Hugging Face has released Quanto, a general-purpose PyTorch quantization backend for the Optimum library.
WebSight: Converting Screenshots into HTML
A large-scale synthetic dataset called WebSight and a model called Sightseer, which convert screenshots into HTML code, have been released.
Accelerating RAG Embedding Performance with Optimum Intel
This covers how to optimize the inference performance of embedding models on Intel CPU environments using Optimum Intel and fastRAG.
Multimodal Reasoning Benchmark 'ConTextual' Released
A new dataset and leaderboard called 'ConTextual' has been released to evaluate multimodal models' ability to jointly reason over text and images.
Hugging Face Supports Community-Driven Dataset Building
Hugging Face and Argilla have unveiled a program that supports building high-quality datasets through community collaboration.
Guide to Running Llama 2 on Intel Gaudi 2
This introduces how to use a text generation pipeline based on Optimum Habana to efficiently run Llama 2 models on Intel Gaudi 2 accelerators.
StarCoder2 and The Stack v2 Released
BigCode has released StarCoder2, a next-generation open-source code LLM, along with The Stack v2, a large-scale code dataset.
HuggingFace Releases TTS Arena
HuggingFace has launched TTS Arena, which compares the quality of TTS models based on human preferences.
A Guide to Watermarking Technology for AI-Generated Content
This summary outlines watermarking, data poisoning, and digital signing technologies used to verify the authenticity of AI-generated content.
Matryoshka Embedding Technology
This introduces Matryoshka Embedding technology, which concentrates key information in the earlier dimensions to reduce embedding dimensionality without performance loss.
LLM Red Teaming Resistance Leaderboard Released
Haize Labs has released a red teaming resistance benchmark that measures LLM vulnerability to sophisticated human attacks.
Hugging Face Releases PEFT Guide for Gemma Models
Hugging Face has released a guide on using PEFT and QLoRA to efficiently fine-tune Gemma models.
Google Unveils Open LLM 'Gemma'
Google has released Gemma, an open LLM family built on Gemini technology.
Upstage Unveils Open Ko-LLM Leaderboard
Upstage has launched the Open Ko-LLM Leaderboard to fairly evaluate the performance of Korean LLMs.
PEFT Updates LoRA Adapter Merging Feature
Hugging Face's PEFT library now supports new merging methods that can efficiently combine LoRA adapters.
Building Custom Models Using Synthetic Data
This presents a method for building low-cost, high-efficiency custom models that achieve GPT-4-level performance by leveraging synthetic data generated by LLMs.
AMD Hosts AI Developer Contest
AMD and Hugging Face are hosting an AI developer contest with a total prize pool of $160,000.
Hugging Face Launches OpenAI-Compatible Messages API
Hugging Face has launched a Messages API compatible with the OpenAI API to support transitions to open LLMs.
SegMoE: MoE Framework for Diffusion Models Released
SegMoE, a Mixture-of-Experts (MoE) framework for diffusion models, has been released.
NPHardEval Leaderboard for LLM Reasoning Evaluation Released
The NPHardEval leaderboard, which dynamically evaluates LLMs' logical reasoning ability using complexity classes, has been released.