— Concepts
Inference
The process of running a trained AI model to get a prediction or output.
What is Inference?
Inference is what happens when you call an LLM — the model is run to produce an output for your input. Distinct from training (where the model is being created). 'Inference cost' refers to the per-call cost of running the model. Companies like Groq specialise in fast inference.
— Related
Terms connected to Inference
Models
LLM (Large Language Model)
An AI model trained on huge amounts of text that can understand and generate human language.
Open →Tools
Groq
A high-speed AI inference provider — 10x faster than typical APIs.
Open →Concepts
Tokens
The chunks of text LLMs process — roughly 0.75 words each.
Open →Infrastructure
API
An interface for programs to call AI models or services programmatically.
Open →— Apply this
From definitions to deployed projects.
Knowing what a term means is step one. ONROL's AI Generalist track gets you shipping projects that use it.
Reserve Free Masterclass