A lightweight Python-based benchmarking suite to evaluate and compare multiple Large Language Models (LLMs) locally using Ollama. This tool measures model speed, latency, token generation performance, resource usage, and response quality, ranking models to help you select the best LLM for your hardware.

A robust Python-based module for identifying and profiling customers across multiple merchants using transactional metadata, behavioral patterns, and similarity search. Ideal for fraud detection, personalization, and unified customer insights in payment ecosystems.

GitMind-MCP is an AI-powered middleware that connects an Ollama-based LLM with the GitHub API. It enables intelligent interaction with GitHub repositories by leveraging natural language understanding and automated API calls.

This project automatically generates a professional PowerPoint presentation based on user input from a simple form. It uses AI to understand the content, corrects grammar and improves the English, and adds relevant images that match the text on each slide.

A desktop application that allows you to chat with any website using AI vision capabilities. The assistant can analyze website screenshots and respond to your questions about the content. The project uses Vision Language Model.

A simple web-based interface for interacting with Ollama models. Includes features like Model Selection, Chat Interface, Dark Theme, and more.

Generate Road, Rider and Vehicle Analytics just by using the sensors of your Smart Phone and using AI/ML capabilities. It uses sensors like Accelerometer. The project also contains RNN-LSTM and RFC Model trainend on self-collected custom data. The repo contains dataset, with all project development asset.

FileNet is a library to visulaize your directories and files with their hierarchical relation in 2D, 3D or JSON. The Library can be found on official PyPI website and installed using 'pip' command 'pip install FileNet'.

The project provides live Traffic Informations with dynamic Traffic Light Control. The repo has Yolo(5su, 8n) and SAM(Segment Anything Model)(sam_b, sam2_b) models used for detection and segmentation tasks.

The project lets you convert any Indian Language Audio to Text using ASR Model.

The project lets you convert any Indian Language Text to Speech using TTS Model. The generated text is also in the Indic Script (eg :Tamil Text -> Tamil Speech)

The project allows you to do Document Retrieval using VLM. Take .pdf file or any file and transform it to image and perform QnA.

Project Components :
Client : It is the frontend that allows user to interact with system.
Indic-ASR : It is a ASR model that transcribes auido input to text in spoken language.
LLM : It is Language Model to generate text output for input from ASR model.
Indic-TTS : It is a TTS model which takes the LLM output as input and generates audio out of it with KN accent.

A collection of terminal-based UI applications built with the Textual framework for Python. This repository contains examples and projects to help you learn how to create rich, interactive terminal applications.