AI Foundations
Architecture
The overall design of a model, including how its layers or components are arranged.
Example
Two language models use different arrangements of layers even though both predict text.
Why it matters
It gives builders ways to suit different tasks and computing limits.
You'll hear it
“Both systems write text, but they're built differently.”
Can you control it?
Set by the team
The people building or running the AI choose this setup. An everyday user generally needs their help to change how this part works.
Common questions
- Can the same design be trained more than once?
- Yes. Different teaching examples or training choices can produce different systems from the same overall design.
- Is architecture the same as model size?
- No. Size describes how much a system contains; architecture describes how its parts are arranged.
- Why do different architectures exist?
- Different designs can suit language, images, sequences, speed, memory, or other requirements.