Concepts

Understand how the INFER platform works under the hood.

Core Architecture

INFER is a decentralized inference marketplace connecting AI compute buyers with GPU node operators. Learn about the platform’s architecture, how requests are routed, and how the token-based billing system works.

Topics

• Architecture — Platform components, data flow, and deployment topology
• Inference Routing — How requests are matched to the best available node
• Billing & Tokens — Token-based payments, pricing models, and settlement