Quantization Engineering

GGUF, AWQ, model compression

Author: AI Learning Club Team · Difficulty: Advanced · Duration: 2 hr 10 min · Modules: 25

Modules

1. Introduction to Quantization Engineering (60 min)
2. Fundamentals of Model Compression (60 min)
3. GGUF: Grouped Quantization Techniques (60 min)
4. GPTQ: Gradient-based Quantization (60 min)
5. AWQ: Adaptive Weight Quantization (60 min)
6. INT4 and INT8 Quantization (60 min)
7. bitsandbytes Library Overview (60 min)
8. Practical Implementation of GGUF (60 min)
9. Practical Implementation of GPTQ (60 min)
10. Practical Implementation of AWQ (60 min)
11. INT4 and INT8 in Practice (60 min)
12. Using bitsandbytes for Quantization (60 min)
13. Quantization-Aware Training (60 min)
14. Post-Training Quantization (60 min)
15. Quantization for Different Model Architectures (60 min)
16. Benchmarking Quantized Models (60 min)
17. Performance Metrics for Quantized Models (60 min)
18. Trade-offs in Quantization (60 min)
19. Advanced Topics in Model Compression (60 min)
20. Case Studies in Quantization Engineering (60 min)
21. Quantization for Edge Devices (60 min)
22. Quantization and Model Interpretability (60 min)
23. Future Directions in Quantization (60 min)
24. Project: End-to-End Quantization Pipeline (60 min)
25. Resources & References (2 min)

Frequently Asked Questions

Related Courses

• MLOps & Model Deployment — Advanced
• Local LLM Architecture — Advanced
• Production Inference — Advanced
• LLM Fine-Tuning — LoRA, QLoRA, PEFT, Instruction Tuning, RLHF, DPO, Evaluation — Advanced

← All courses