← Back to Mochi’s learning shelfMochi’s Minute · AI basics
What happens when AI answers?
Inference is a trained model doing work on a new input.
By Tony Yu / tkal · October 31, 2026 · One-minute visual guide
Learn it. Then use it.
01 · TRAINING
Learn the recipeLearn patterns
from training data.
→02 · INFERENCE
Use the recipeApply the trained model
to a new input.
Keep this bitInference is the use of a trained AI model to generate an output from new inputs.
For a chatbot
A new prompt goes in. The trained model generates an answer.
Why care?
Answering still uses computing resources after training is finished.
Open the original sources
Source links open separately. Mochi’s pictures simplify the idea; the original documents carry the detail.
