Create an AI squad to automate literally anything on your laptop without GPU
—
In this article, I will show how to locally build Agents, which are the peak of Generative AI (GenAI) state-of-the-art, explaining the difference from normal LLM.
Generally speaking, Intelligent Agents are entities acting with enough understanding to perceive the environment, take actions autonomously to achieve specific goals, and improve by acquiring knowledge (like humans).
In tech, an AI Agent is an autonomous program that performs tasks, makes decisions, and communicates with other entities. Usually, Agents are provided with a set of tools that they can decide to use while completing the task. This concept extends Reinforcement Learning, where an agent picks from a defined set of actions to maximize the reward in an interactive environment.
Large Language Models (LLM) are not Agents. LLMs are Neural Networks that utilize word embedding and transformer architecture to perform advanced NLP. They have acquired a great comprehension of the human language, but don’t perform any action outside their knowledge base.
In GenAI, Agents are AI systems designed to process sequential reasoning, with the option of executing external tools (i.e. database query, web…