Definition

Inference in LLMs

Inference is the stage where a pre-trained and fine-tuned Large Language Model is utilized to generate predictions for new inputs. During this process, the model calculates the probability for each potential output and ultimately selects the one with the highest probability.

0

1

Updated 2026-05-03

Contributors are:

Who are from:

Tags

Ch.5 Inference - Foundations of Large Language Models

Foundations of Large Language Models

Foundations of Large Language Models Course

Computing Sciences

Ch.2 Generative Models - Foundations of Large Language Models