FeedSources
⌘K
HF

Hugging Face Blog

74 articles total

Go to source

We’re on a journey to advance and democratize artificial intelligence through open source and open science

Go to source
  •  Rebuilding AUTOMATIC1111 with Gradio Workflow
  •  IBM releases SOTA Granite Time Series PatchTST-FM-r2 model with commercial-friendly license
  •  Safety for Whom? Refusing the Right Subset of a Topic, Not the Whole Topic
  •  NeoMME: an efficient Multimodal-native and Multilingual Encoder
  •  Fine-tuning a 350M Model for Better Structured Outputs in 100 GRPO Steps
  •  Give Your Coding Agents a Memory You Own
  •  Training a coding model to paint watercolours with TRL and OpenEnv
  •  Real-Time Intelligence with IBM Time Series Models on Confluent
  •  BenchMIRT: What are LLM benchmarks actually measuring?
  •  Introducing @huggingface/kernels: 200+ WebGPU Kernels for Local AI
  •  The Open ASR Leaderboard Adds Its First Global South Language
  •  Training and Finetuning Multi-Vector Embedding Models with Sentence Transformers
  •  Granite 4.2 LLMs: How They're Built
  •  Quantization-Aware Healing: a compressed, 4-bit model that outperforms its full-precision original
  •  How Hugging Face Inference Endpoints, Jobs, and Buckets Power Search on Papers with Code
  •  Wire It, Run It, Deploy It: AI Workflows in Gradio
  •  Measuring benchmark optimization in speech recognition
  •  Up to 3.2x Faster Inference with LFM2.5-DSpark
  •  LFM2.5 Q4\_0 Checkpoints from Quantization-Aware Distillation
  •  How Much Memory Does Your Agent Actually Need?