AI BriefingHuggingFace Blog news
The latest AI and developer news from HuggingFace Blog, with short summaries.
Visit website →Sentence Transformers Training Guide
This guide explains in detail how to fine-tune embedding models using the new training approach introduced in Sentence Transformers v3.0.
LLM Security Evaluation Framework Released
CyberSecEval 2, a framework for comprehensively evaluating LLM security vulnerabilities and cyberattack capabilities, has been released.
TII Releases Falcon 2 11B Model and VLM
TII has released the 11B-scale Falcon 2 LLM and VLM, trained on over 5 trillion tokens.
Hugging Face Supports AWS Inferentia2
Hugging Face now supports AWS Inferentia2, enabling deployment of over 100,000 models on SageMaker and Inference Endpoints.
Hugging Face-Microsoft Announce Expanded AI Collaboration
Hugging Face and Microsoft are strengthening their strategic collaboration, including deploying open models via Azure and optimizing for AMD GPUs.
Hugging Face Supports AMD MI300
Hugging Face has integrated AMD Instinct MI300 GPUs into its platform to support model deployment and inference optimization.
Dell Enterprise Hub Unveiled
Hugging Face has launched Dell Enterprise Hub, which makes it easy to train and deploy open models on-premise via the Dell platform.
Hugging Face Spaces Launches Dev Mode
Hugging Face has launched Dev Mode, which lets you edit Spaces code directly via VS Code and SSH.
Hugging Face Unveils KV Cache Quantization Feature
KV Cache quantization reduces LLM memory usage and enables longer context generation.
HuggingFace Unveils Arabic LLM Leaderboard
The Open Arabic LLM Leaderboard has been released to professionally evaluate the performance of Arabic LLMs.
HF-LangChain Integration Package Released
Hugging Face and LangChain have partnered to release the `langchain-huggingface` package, which supports the latest features.
Google Unveils Open VLM PaliGemma
Google has released PaliGemma, a new open vision language model (VLM) that combines SigLIP and Gemma.
Hugging Face Releases smolagents Library
Hugging Face upgraded Transformers Agents into an independent library called smolagents and released it.
Guide to Building RAG Based on Intel Gaudi 2
This introduces how to build cost-efficient enterprise RAG applications and their architecture using Intel Gaudi 2 and Xeon CPUs.
Hugging Face Supports AWS Marketplace Subscription
Hugging Face has launched the ability to subscribe to Enterprise Hub through AWS Marketplace.
Open Leaderboard Dedicated to Hebrew LLMs Released
A new open leaderboard that evaluates LLM performance by reflecting the linguistic characteristics of Hebrew has been released.
Hugging Face Introduces LLM Leaderboard
Artificial Analysis has introduced an LLM performance leaderboard on Hugging Face that comprehensively evaluates price, speed, and quality.
ASR and Speaker Diarization Based on HF Endpoints
It introduces how to implement a custom handler that combines Whisper and Pyannote in Hugging Face Inference Endpoints to support speaker diarization and speculative decoding.
LLM Evaluation Volatility Based on Prompt Format
A study has found that minor formatting changes in prompts significantly affect LLM benchmark scores and model rankings.
StarCoder2-15B-Instruct Released
A code generation model that trained itself using StarCoder2-15B, without human intervention or paid LLMs, has been released.