HacxGPT Jailbreak 🚀: Unlock the full potential of top AI models like ChatGPT, LLaMA, and more with the world's most advanced Jailbreak prompts 🔓.

Karbon is a desktop app that uses AI to generate and iteratively refine HTML/CSS/JS websites from simple prompts. Built with Python and Tkinter, it provides a live preview and export functionality.

SAM3: Segment Anything Model 3 is an experimental, multi-functional computer vision application that provides a unified interface for text-conditioned and click-interactive segmentation. Built around Meta's advanced facebook/sam3 multi-model architecture.

sauraya AI is a Mobile interface that allows you to interact with large open source artificial intelligence models.

This project demonstrates three progressively advanced approaches to converting unstructured data into structured formats using AI

In this repository, I have explained the working of the LLaMA Model, provided the code for building it from scratch, and demonstrated how to train it.

Node.js REST api to chat with open source ai models running at local machine using ollama and langchain.

This demonstrates the process of adapting a large scale pretrained model, MetaCLIP 2, for fine tuning a specific downstream task: image classification.

Leveraging Llama3.2 LLM to generate problems based on lecture/class content; Prototype for integration into larger application

AI services

Protein Structure prediction and Analysis -Model Based on Meta's ESMFold

Future Domain Extensions: Shaping the Internet in 2030

AIPROJECTPAL is an AI assistant designed to help users overcome challenges in project implementation. It offers solutions, code generation, and knowledge retrieval across various domains. Developed by Team AIPROJECTPAL, led by Herish Chaniyara, it leverages advanced AI models to provide innovative solutions and continuous learning.

LLama2 Chat assistance with streamlit for frontend and hosted on replicate.

"Flipping Metaverse Domains for Profit"

Star Stream: A powerful, custom-built streaming platform for immersive event broadcasting. Features include live video streaming, audio transcription, AI-powered video Q&A, and real-time chat – all in one seamless solution.

The project aims to democratize access to information by enabling transparent and inclusive AI interactions.

A lightweight Streamlit app that generates blog content using a locally hosted, quantized LLaMA 2 model.

Implementation of a Question Answering System using LLMs

This Streamlit-based system supports multi-format document(PDFs, DOCX, HTML, and images) ingestion, text extraction, chunking, and FAISS-based indexing, enabling question answering via a lightweight, API-free Meta/LLaMA-compatible inference interface