Intent-classifier-model is a lightweight machine learning tool designed to categorize short customer support text into specific intents such as greeting, question, complaint, praise, or other. The project demonstrates the end-to-end workflow of training a small text classification model, serializing the model artifact to disk, and deploying it as a production-ready service. Users can set up a local virtual environment, install dependencies, and execute the training script to generate the model file. The software includes a Flask-based API server that exposes a predict endpoint for real-time inference. Developers can send JSON payloads containing customer messages via HTTP POST requests to receive the predicted intent class along with confidence probabilities for each category. This solution is ideal for integrating automated intent detection into chatbots, helpdesk systems, or automated routing pipelines without requiring heavy computational resources.