Zen-AI is a full-stack AI chat app built with React, Node.js, and Express. It uses Mongo DB, Pinecone vector search and Gemini embeddings for context-aware AI responses, supports user authentication, and is fully deployed with SPA routing and secure APIs.
Playing with Generative AI
A full-stack AI chatbot built with React and Node.js, powered by Google's Gemini API. Supports normal conversational chat with streaming responses, and lets you upload a PDF to ask questions grounded in its content using Retrieval-Augmented Generation (RAG).
A test series generator for JEE-Mains using RAG and LLM
CacheRAG — An efficient RAG optimization pipeline using exact caching, semantic caching, and MMR reranking to reduce LLM latency, token usage, and redundant AI generation costs.
A Node.js model tool, which supports embedding with batched input, parallel execution, isolated child-process workers (default) or in-process threads, quantization, optional GPU acceleration, and Hugging Face auth.
Retrieval Augmented Generation POC.
A Chat AI Application Using Retrieval-Augmented Generation (RAG) and PineCone as Vector DB.
To associate your repository with the embedding-vectors topic, visit your repo's landing page and select "manage topics."