LLM
4 articles
-
Small Language Models vs Large Language Models: Is Smaller Better?
Small language models vs large language models: accuracy, cost, efficiency and the memory and hardware each needs, with a practical sizing table.
Wei Fei
·
-
AI Model Distillation: Can a Smaller Model Match a Larger One?
Model distillation explained: how a small student model learns from a large teacher, how close it gets, the limits, licensing and hardware needs.
Wei Fei
·
-
What Is KV Cache? Why Long AI Conversations Need More GPU Memory
KV cache and context windows explained with worked memory maths for 8B and 70B models, multi-user sizing, and ways to cut KV memory.
Wei Fei
·
-
Mixture of Experts (MoE) Models: Why Bigger Doesn't Always Mean More Expensive to Run
Mixture of experts explained: total vs active parameters, why MoE models need lots of memory but little compute per token, and hardware sizing.
Wei Fei
·
Wei Fei
