Introduction

mkinf comprises two core components:

• Library of Hosted Agents: A collection of pre-built agents that streamline the development process by providing ready-to-use functionalities.
• Scalable GPU Infrastructure: An on-demand GPU resource management system that efficiently supports AI model inference, dynamically adjusting to meet the demands of a variable user base.

This documentation details the features and technical aspects of mkinf to help you integrate and optimize its capabilities in your projects.

​

mkinf hub

An open-source library of hosted AI agents and tools that developers can integrate into their graph frameworks with a simple SDK or API call. This accelerates both development and deployment.

​

mkinf grid

Distributed inference from a single endpoint.