LangGraph: Build a Multifunctional Robot with qwen2.5 for Private Deployment

Build an agent-based chatbot using qwen2.5 to gather user inputs, generate prompts, and switch states dynamically for optimized functionality.

Building a robot using an agent-based approach is quite fun, so I'd like to share it with you all.

In this example, we will create a chatbot that helps users generate prompts.

The bot will first gather requirements from the user, then generate a prompt (and refine it based on further input).

These steps are split into two states, with the LLM deciding when to switch between them. The graphical representation of the system can be seen below.

Gathering Information

First, let's define the part of the diagram that collects user requirements.

This will be an LLM call with a specific system message.

It will have access to a tool that can be invoked when it's ready to generate the prompt.