Skip to content

Vivid Labs AI

Now Watching

How LLMs Use Tools

How LLMs Use Tools /Tool Binding and the Runnable Contract

Tool Binding and the Runnable Contract

Understand how tools are bound to a chat model and how LangChain's Runnable contract keeps inputs, outputs, and composition predictable.

Queue

7 videos in this course

How LLMs Use Tools

Jump across the full chapter sequence without leaving the player.

01

Why Language Models Need Tools

See where language models reach their limits and how external tools add current data, calculations, and real-world actions to an LLM workflow.

02

Tool Binding and the Runnable Contract

Understand how tools are bound to a chat model and how LangChain's Runnable contract keeps inputs, outputs, and composition predictable.

03

The Tool-Calling Lifecycle

Follow a tool request from model output through execution and back into the conversation so the model can use the result correctly.

04

Built-In and Custom LangChain Tools

Compare ready-made LangChain tools with custom tools and learn how names, descriptions, and schemas guide reliable model behavior.

05

Binding Tools and Planning an Application

Turn a product goal into a tool-enabled application plan, then bind the right capabilities to the model that will coordinate them.

06

Geocoding and Weather Tool Design

Design a realistic multi-step tool flow that resolves a place, retrieves weather data, and passes clean structured results between calls.

07

Multi-Tool Orchestration and Message History

Coordinate several tools in one workflow while preserving the message history the model needs to reason across each step.