Back to glossary
261 terms · 10 chapters
LLM Optimization Dictionary
261 terms covering pre-training, fine-tuning, quantization, inference, decoding, architecture, and serving optimization for large language models.
35 terms
Pre-Training Optimization
39 terms
Fine-Tuning Optimization
22 terms
Quantization
39 terms
Inference Acceleration
16 terms
Decoding Strategies
34 terms
Architecture Optimization
20 terms
Compression Techniques
16 terms
Context & Memory Optimization
22 terms
Serving & Systems Optimization
18 terms
Evaluation & Efficiency Metrics