Local Low-Code Chatbot with Ollama and Flowise

Create a fully functional, local chatbot using Flowise and Ollama. This project is designed to provide a low-code, privacy-friendly solution for building intelligent conversational bots. 🚀 This was made as part of a blog that you can find here

Features

• Low-Code Workflow: Build chatbots visually without heavy coding.
• Local Hosting: Keeps your data private and secure.
• Customisable: Fully adjustable to your needs.
• Powered by Open-Source Models: Utilizes Ollama's LLMs for AI capabilities.

Workflow Overview

Core Components

1. ChatOllama: Provides AI responses using Ollama's models.
2. Buffer Memory: Retains chat history for continuity.
3. Conversation Chain: Integrates ChatOllama and Buffer Memory to enable interactive conversations.

Getting Started

Prerequisites

• Docker (for running Ollama locally)
• Node.js and npm (for Flowise)
• A compatible LLM model (e.g., SARA-llama3.2)

Installation

1. Clone the repository:

   git clone https://github.com/dwain-barnes/local-low-code-chatbot-ollama-flowise.git
   cd local-low-code-chatbot-ollama-flowise

2. Start Ollama:

   docker run -p 11434:11434 ollama/server

3. Install Flowise:

   npm install -g flowise
   flowise start

4. Import the provided Flowise JSON (basic-local-chatbot-flowise.json) into Flowise.

Running the Chatbot

1. Access the Flowise editor at http://localhost:3000.
2. Load and activate the workflow.
3. Start chatting with the bot in the interface.