Community Blog & Articles

Community Articles

Microsoft and Hugging Face expand collaboration

Falcon-Edge: A series of powerful, universal, fine-tunable 1.58bit language models.

The Transformers Library: standardizing model definitions

Improving Hugging Face Model Access for Kaggle Users

Blazingly fast whisper transcriptions with Inference Endpoints

Vision Language Models (Better, faster, stronger)

LeRobot Community Datasets: The “ImageNet” of Robotics — When and How?

How to Build an MCP Server with Gradio

The 4 Things Qwen-3’s Chat Template Teaches Us

Welcoming Llama Guard 4 on Hugging Face Hub

Introducing AutoRound: Intel’s Advanced Quantization for LLMs and VLMs

PipelineRL

Tiny Agents: an MCP-powered agent in 50 lines of code

Finetuning olmOCR to be a faithful OCR-Engine

NEW Articles from Team or Enterprise organizations will get promoted to the main section.

Community Blog & Articles

UltraBERT: One Encoder for Classification and Retrieval

The Missing Piece in LLM Memory: Why Long Context and Retrieval Aren't Enough for Belief Revision

Tokenmaxxing BERT: 50x faster tokenization

Impactful scheduling for GPU clusters

Apertus 1.5 Release Notes

Introducing Lumma-Fev-27B

One Decision Model Instead of a Ranking Team: Short-Video and News Recommendation with autotrust/JEV-27B-VL

ColGrep-Agent: Cutting Frontier Coding Agents’ Tokens by a Third

Falcon OCR Arabic: 270M Parameters State-of-the-Art Arabic OCR

When agents self-improve, what should persist?

UIBK-Embedding-v1: A Dense and a Late-Interaction Retriever in One Model, One Index and One Search

Backpropagation on an Inference-Only NPU: Pretraining VKORE-46M-NPU From Scratch

Fine-Tune ASR for Your Domain with NVIDIA Speech Agent Skills

LightOnOCR-3: High-Performance OCR and Layout Extraction in One Model

Falcon-Emirati: When an LLM Learns the Dialect, the Culture, and the Nuance

New in llama.cpp: Decision Models

Sparseup-fr: A French replication of Linkup's sparse embedding model

Introducing Llama Windows: llama.cpp in your Windows tray

One Model Family, Two Gold-Level Results: Fine-Tuning Nemotron for IOI and IMO

Multimodal open d1 decision models for the edge

Microsoft and Hugging Face expand collaboration

Falcon-Edge: A series of powerful, universal, fine-tunable 1.58bit language models.

The Transformers Library: standardizing model definitions

Improving Hugging Face Model Access for Kaggle Users

Blazingly fast whisper transcriptions with Inference Endpoints

Vision Language Models (Better, faster, stronger)

LeRobot Community Datasets: The “ImageNet” of Robotics — When and How?

How to Build an MCP Server with Gradio

The 4 Things Qwen-3’s Chat Template Teaches Us

Welcoming Llama Guard 4 on Hugging Face Hub

Introducing AutoRound: Intel’s Advanced Quantization for LLMs and VLMs

PipelineRL

Tiny Agents: an MCP-powered agent in 50 lines of code

Finetuning olmOCR to be a faithful OCR-Engine

UltraBERT: One Encoder for Classification and Retrieval

The Missing Piece in LLM Memory: Why Long Context and Retrieval Aren't Enough for Belief Revision

Tokenmaxxing BERT: 50x faster tokenization

Impactful scheduling for GPU clusters

Apertus 1.5 Release Notes

Introducing Lumma-Fev-27B

One Decision Model Instead of a Ranking Team: Short-Video and News Recommendation with autotrust/JEV-27B-VL

ColGrep-Agent: Cutting Frontier Coding Agents’ Tokens by a Third

Falcon OCR Arabic: 270M Parameters State-of-the-Art Arabic OCR

When agents self-improve, what should persist?

UIBK-Embedding-v1: A Dense and a Late-Interaction Retriever in One Model, One Index and One Search

Backpropagation on an Inference-Only NPU: Pretraining VKORE-46M-NPU From Scratch

Fine-Tune ASR for Your Domain with NVIDIA Speech Agent Skills

LightOnOCR-3: High-Performance OCR and Layout Extraction in One Model

Falcon-Emirati: When an LLM Learns the Dialect, the Culture, and the Nuance

New in llama.cpp: Decision Models

Sparseup-fr: A French replication of Linkup's sparse embedding model

Introducing Llama Windows: llama.cpp in your Windows tray

One Model Family, Two Gold-Level Results: Fine-Tuning Nemotron for IOI and IMO

Multimodal open d1 decision models for the edge