HuggingFace Courses

Introduction to Multimodal AI with Hugging Face

Build multimodal pipelines: caption products with CLIP and VLMs, extract invoice fields, transcribe calls with Whisper, generate images with Diffusers, and add smolagents, MCP and safety filtering.

• Coursera
• 4 weeks, 1 hour a week
• Self-Paced
• Paid Course

Getting Started with Hugging Face Transformers

Navigate the Hugging Face Hub's two million models: run pipeline inference on text, vision and audio, tokenize with AutoTokenizer, load AutoModel classes, and judge model cards.

• Coursera
• 4 weeks, 1 hour a week
• Self-Paced
• Paid Course

Building AI Apps with Hugging Face Spaces and Gradio

A model stuck in a notebook is a model nobody uses: fine-tune with the Trainer API, build Gradio interfaces and chatbots, and deploy to Hugging Face Spaces.

• Coursera
• 4 weeks, 2 hours a week
• Self-Paced
• Paid Course

Building Generative AI Applications with Gradio

Build interactive generative AI demos with Gradio, including text summarization, image captioning, image generation, and open-source LLM chat.

• Independent
• 1 hour
• Self-Paced
• Free Online Course

Food for Diffusion

Explore how diffusion models generate textures compositionally and how fixed- or dynamic-size training affects image composition.

• YouTube
• 34 minutes
• Self-Paced
• Free Video

Stable Diffusion and Friends - High-Resolution Image Synthesis via Two-Stage Generative Models

A research-focused tour of Stable Diffusion’s two-stage latent generative architecture, from image compression and diffusion modeling to text-to-image synthesis.

• YouTube
• 28 minutes
• Self-Paced
• Free Video

Beyond Text - Giving Stable Diffusion New Abilities

A technical deep dive into adapting Stable Diffusion beyond text prompts for image variations, face control, super-resolution, and video prediction.

• YouTube
• 31 minutes
• Self-Paced
• Free Video

Few-Shot Learning in Production

Train and deploy few-shot language models with SetFit, then compress them through knowledge distillation and quantization for efficient inference.

• YouTube
• 1 hour 22 minutes
• Self-Paced
• Free Video

Reinforcement Learning from Human Feedback - From Zero to ChatGPT

An overview of how human feedback, reward models, and reinforcement learning align large language models such as ChatGPT.

• YouTube
• 1 hour 1 minute
• Self-Paced
• Free Video

Reinforcement Learning from Human Feedback - From Zero to ChatGPT

Explore how human feedback, reward models, and PPO align large language models such as ChatGPT.

• YouTube
• 1 hour 1 minute
• Self-Paced
• Free Video

XLS-R: Large-Scale Cross-Lingual Speech Representation Learning on 128 Languages

A research talk on XLS-R’s self-supervised cross-lingual speech representations, trained on 128 languages and applied to recognition, translation, and classification.

• YouTube
• 27 minutes
• Self-Paced
• Free Video

Hugging Face Workshops - Pretraining Language Models & CodeParrot

Learn how to pretrain a causal language model for code generation by building and evaluating CodeParrot.

• YouTube
• 58 minutes
• Self-Paced
• Free Video

Managed Training with Amazon SageMaker and Transformers

Train and automatically publish a Transformer text-classification model to the Hugging Face Hub using managed Amazon SageMaker jobs.

• YouTube
• 25 minutes
• Self-Paced
• Free Video

Building Machine Learning Applications Fast

Learn to use Gradio to quickly turn machine learning models into interactive, shareable applications.

• YouTube
• 23 minutes
• Self-Paced
• Free Video

Get Your Own Tokenizer With Transformers and Tokenizers

Build and train a custom tokenizer for language models with Hugging Face Transformers and Tokenizers, including vocabulary, special tokens, and post-processing.

• YouTube
• 23 minutes
• Self-Paced
• Free Video