LLM.

The LLM is the brain. It stands for Large Language Model, and it's just the Chat model, Claude or ChatGPT or Gemini, sitting at the center doing the thinking.

When people say "the AI," they usually mean the LLM. It's the part that reads, reasons, and writes.

An AI agent is that same brain with a body bolted on. The LLM thinks, and the Agent harness, tools, and context around it give it hands so it can go and do the work. When a setup feels smart or dumb, that's usually the model underneath. When it feels capable or useless, that's almost always the body around it.

One thing worth knowing if you're building: you can usually swap the brain. The same harness runs on Claude, GPT, or a Local model, so you pick the model that fits the job and the rest of your setup stays put.