AI BriefingHuggingFace Blog news
The latest AI and developer news from HuggingFace Blog, with short summaries.
Visit website →How to Use the PatchTST Time-Series Model
This explains the structure of PatchTST, a model that improves time-series forecasting performance, and how to use it with the Hugging Face library.
TGI Adds Support for AWS Inferentia2
Hugging Face's LLM serving solution TGI is now officially supported on AWS Inferentia2 and Amazon SageMaker.
HuggingFace releases CAI recipe for open LLMs
HuggingFace has released an implementation methodology, related tools, and datasets for **Constitutional AI** for open source models.
Launch of a Practice-Focused LLM Evaluation Leaderboard
Patronus AI has launched a new LLM leaderboard that evaluates real-world enterprise use cases such as finance and law.
StarCoder Inference Accelerated 7x on Intel Xeon
Through Optimum Intel and quantization techniques, StarCoder's inference performance on Intel Xeon processors was improved by more than 7x.
LLM Hallucination Measurement Leaderboard Released
An open-source leaderboard for evaluating factuality and faithfulness hallucinations in LLMs has been released.
LLM Reliability Evaluation Leaderboard Released
A new Safety Leaderboard based on DecodingTrust, which comprehensively evaluates LLMs' toxicity, bias, security, and more, has been released.
Hugging Face and Google Cloud Announce Strategic Partnership
Hugging Face and Google Cloud have entered into a strategic partnership to support building AI based on open models.
Open-Source LLMs Prove Their Agentic Capabilities
Open-source LLMs like Mixtral have reached a level where they can perform ReAct-style agent workflows.
IBM PatchTSMixer released on HF
IBM Research's lightweight time series model PatchTSMixer has been integrated into the HuggingFace Transformers library.
Implementing Low-Resource ASR with Wav2Vec2-BERT
This covers how to use Meta's Wav2Vec2-BERT to improve automatic speech recognition (ASR) performance for languages with limited data.
Comparing DPO, IPO, and KTO Preference Alignment Methodologies
An analysis compared the performance and characteristics of major LLM preference alignment algorithms including DPO, IPO, and KTO.
SDXL Turbo Inference Speed Improved by Up to 229%
Inference performance of SD Turbo and SDXL Turbo models was significantly improved through ONNX Runtime and Microsoft Olive.
How to Freely Deploy ComfyUI Workflows
This explains how to convert ComfyUI workflows into Python, turn them into a Gradio app, and deploy them for free using Hugging Face ZeroGPU.
Hugging Face Releases Guide for Building Leaderboards
It introduces the technical method of building a custom LLM leaderboard using Hugging Face's new template, along with Vectara's case study.
Unsloth Doubles LLM Fine-Tuning Speed
The Unsloth library can speed up LLM fine-tuning by up to 2x while reducing memory usage by 40%.
Non-Diffusion Image Generation Model aMUSEd Released
aMUSEd, an efficient non-diffusion text-to-image generation model that recreates Google's MUSE, has been released.
SDXL LoRA Training Optimization Script Released
Hugging Face has released a new script that maximizes SDXL Dreambooth LoRA training performance.
Technique Doubles Whisper Inference Speed
By applying Speculative Decoding, the inference speed of Whisper was doubled while maintaining accuracy.
2023 Open LLM Trends and Technical Summary
This summarizes the development process of open source LLMs in 2023 and their core technical components.