Distillery AI Guide

Distillery AI Guide is a comprehensive, interactive course on model distillation, the machine-learning technique that compresses large 'teacher' models into smaller 'student' models without losing most of their intelligence.

WHAT YOU'LL LEARN
Across 8 in-depth learning modules you will master:
- Introduction to Model Distillation - what it is, history, and key terminology
- Why Distill: Efficiency & Deployment - latency, memory, energy, and total cost of ownership
- Knowledge Distillation Fundamentals - the teacher-student loss, soft targets, and gradients
- Temperature & Soft Labels - how the temperature knob and softened probabilities transfer knowledge
- Data-Free & Feature Distillation - distilling without the original data and from intermediate layers
- Self-Distillation & Teacher-Student Variants - self, ensemble, mutual, online, and cross-architecture distillation
- Quantization, Pruning & Compounding - combining distillation with the other major compression techniques
- Workflows, Tools & Evaluation - production pipelines, tooling, and release-ready evaluation

HOW YOU LEARN
- 8 modules with over 60 detailed, code-rich lessons and rich HTML explanations
- 5 quizzes with 125 questions to test and reinforce your understanding (pass score 80%)
- A 100-term glossary covering distillation, compression, and deployment vocabulary
- Practical recipes, PyTorch-style code walkthroughs, and real-world case studies

WHO IT'S FOR
Machine-learning engineers who want cheaper, faster models; researchers exploring model compression; and students learning neural-network optimization and deployment. All content is original, factual, and educational.

DISCLAIMER: This is an unofficial educational guide and is not affiliated with or endorsed by any distillation tool vendor.

Recent changes:
Version 1.0 - Initial release of Distillery AI Guide.

What's included:
- 8 comprehensive learning modules (64 lessons) on model distillation
- 5 quiz sets with 125 self-assessment questions, pass score 80%
- 100-term glossary of distillation and model-compression terminology
- PyTorch-style code examples and step-by-step recipes
- Real-world case studies across edge, on-device, server, and large-language-model deployment

All content is original and intended for educational purposes only.