Become an AI engineer

Model optimization

Model optimization is a unit of 9 lessons in Rabbithole's Become an AI engineer course, at levels 3 to 5.

Lessons

1. Hosted fine-tuning: what it is, what it costs, how to steer itLevel 3
2. Model export and formatsLevel 3
3. Model Merging: Many Models, One Set of WeightsLevel 4
4. Parameter-efficient fine-tuning: train a little, keep the restLevel 4
5. Post-training quantisation: splitting the difficultyLevel 4
6. Numeric precision formats: bits, ranges and what to compute inLevel 4
7. Preference optimisation in practiceLevel 4
8. Pruning and sparsity: removing weights without breaking the modelLevel 5
9. Quantisation-aware trainingLevel 5

Sources

These lessons draw on 26 sources. The most used:

• Optimum documentation
• Scaling Laws for Precision
• PEFT documentation
• AWQ: Activation-aware Weight Quantization for LLM Compression and Acceleration
• DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligence
• DeepSpeed — Getting Started and Tutorials
• SmoothQuant: Accurate and Efficient Post-Training Quantization for Large Language Models
• SparseGPT: Massive Language Models Can Be Accurately Pruned in One-Shot

All units in Become an AI engineer