🏛️Inference Provider
Also known as AI Inference Provider, Model Host, Inference API
A firm that runs AI models for you. You pay each time you use one.
Think of it like
It is like a bakery with a big oven. You need no oven at home.
Example
Emma's app uses an open Llama model. She owns no graphics card, so she sends each prompt to an inference provider and pays per token.
Why it matters
Open weights are free, but the computer to run them is not. A provider turns that cost into a small bill for each use.
Definition
Common mistake
Key takeaways
- 1Their chips. You need no hardware.
- 2Pay per use. By the token or the second.
- 3Many models. Mostly open ones.