Inference
The moment when an AI model actually runs and produces an answer or prediction based on the prompt you gave it.
In Plain English
Inference is the AI's performance time. Training the AI is like a student studying for months, while inference is the student actually taking the test and answering a question. Every time you hit send on ChatGPT and watch the text appear, you are watching inference happen. It requires a lot of computing power, which is why AI companies have massive data centers.
Real-World Example
The three seconds it takes for an AI to generate a picture of a sunset after you type the prompt.