Understanding Agentforce: Reasoning Engine vs. LLM

On the Agentforce Specialist exam, it’s critical to understand the difference between the reasoning

engine and the large language model, and how they work together during an agent interaction.

Salesforce tests this distinction often — and subtly. Let’s break it down in exam language.

First, the reasoning engine. The reasoning engine — also called the planner service

— is responsible for orchestrating the entire interaction. It decides: Which topics to

launch Which actions to run And the order those actions should execute

In other words, the reasoning engine controls what happens. If the exam mentions orchestration,

planning, or launching actions, they are pointing you to the reasoning engine.