Community Blog & Articles

Community Articles

🏟️ Smol AI WorldCup: A 5-Axis Benchmark That Reveals What Small Language Models Can Really Do

Beyond Semantic Similarity: Introducing NVIDIA NeMo Retriever’s Generalizable Agentic Retrieval Pipeline

How NVIDIA AI-Q Reached \#1 on DeepResearch Bench I and II

NEO-unify: Building Native Multimodal Unified Models End to End

Uncensor any LLM with abliteration

How NVIDIA Builds Open Data for AI

Arabic TTS Arena: Ranking Voice Models the Way Chess Ranks Grandmasters

Build an Agent That Thinks Like a Data Scientist: How We Hit #1 on DABStep with Reusable Tool Generation

Expanding the Alpamayo Open Platform for Developing Reasoning AVs Across Models, Data, and Simulation

about 18 hours ago

SILMA TTS: A Lightweight Open Bilingual Text to Speech Model

Norm-Preserving Biprojected Abliteration

Code Concepts: A Large-Scale Synthetic Dataset Generated from Programming Concept Seeds

Code a simple RAG from scratch

KV Caching Explained: Optimizing Transformer Inference Efficiency

From GRPO to DAPO and GSPO: What, Why, and How

We’re open-sourcing our text-to-image model and the process behind it

LLM based Audio models

MARL: Runtime Middleware That Reduces LLM Hallucination Without Fine-Tuning

Pruna 0.3.2: More OSS Algos, More Ways to Optimize

Tokenization is Killing our Multilingual LLM Dream

Holotron-12B - High Throughput Computer Use Agent

The First Healthcare Robotics Dataset and Foundational Physical AI Models for Healthcare Robotics

hubstorageannouncement

Introducing Storage Buckets on the Hugging Face Hub

+8

Keep the Tokens Flowing: Lessons from 16 Open-Source RL Libraries

+5

Granite 4.0 1B Speech: Compact, Multilingual, and Built for the Edge

guidedistributed-trainingaccelerate

Ulysses Sequence Parallelism: Training with Million-Token Contexts

lerobotrobotics

LeRobot v0.5.0: Scaling Every Dimension

+6

Bringing Robotics AI to Embedded Platforms: Dataset Recording, VLA Fine‑Tuning, and On‑Device Optimizations

open-sourcediffusersgenerative

Introducing Modular Diffusers - Composable Building Blocks for Diffusion Pipelines

PRX Part 3 — Training a Text-to-Image Model in 24h!

mixture-of-expertsoptimizationtransformers

Mixture of Experts (MoEs) in Transformers

+3

February 26, 2026

communityopen-sourcellm

GGML and llama.cpp join HF to ensure the long-term progress of Local AI

+2

February 20, 2026

llmfine-tuningtraining

Train AI models with Unsloth and Hugging Face Jobs for FREE

+2

February 20, 2026

IBM and UC Berkeley Diagnose Why Enterprise Agents Fail Using IT-Bench and MAST

February 18, 2026

Community Articles

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

🏟️ Smol AI WorldCup: A 5-Axis Benchmark That Reveals What Small Language Models Can Really Do

Beyond Semantic Similarity: Introducing NVIDIA NeMo Retriever’s Generalizable Agentic Retrieval Pipeline

How NVIDIA AI-Q Reached \#1 on DeepResearch Bench I and II

NEO-unify: Building Native Multimodal Unified Models End to End

Uncensor any LLM with abliteration

How NVIDIA Builds Open Data for AI

Arabic TTS Arena: Ranking Voice Models the Way Chess Ranks Grandmasters

Build an Agent That Thinks Like a Data Scientist: How We Hit #1 on DABStep with Reusable Tool Generation

Expanding the Alpamayo Open Platform for Developing Reasoning AVs Across Models, Data, and Simulation

about 18 hours ago

SILMA TTS: A Lightweight Open Bilingual Text to Speech Model

Norm-Preserving Biprojected Abliteration

Code Concepts: A Large-Scale Synthetic Dataset Generated from Programming Concept Seeds

Code a simple RAG from scratch

KV Caching Explained: Optimizing Transformer Inference Efficiency

From GRPO to DAPO and GSPO: What, Why, and How

We’re open-sourcing our text-to-image model and the process behind it

LLM based Audio models

MARL: Runtime Middleware That Reduces LLM Hallucination Without Fine-Tuning

Pruna 0.3.2: More OSS Algos, More Ways to Optimize

Tokenization is Killing our Multilingual LLM Dream

View all articles