RAG implemented from scratch without using LangChain and LangGraph - designed specifically for processing and querying PDF documents with advanced support for visual content like tables, charts, and mathematical formulas.

使用 LangChain 把 PDF 生成音频播客

RAG-based Multi-PDF Question Answering system with semantic search (FAISS) and LLM inference (Groq & Ollama)

Vision-first document RAG for PDF, DOCX, and image QA/extraction with ColQwen2, Qwen2.5-VL, Qdrant, and FastAPI.

Production-ready multilingual RAG system for scientific PDFs. Supports 10+ Indic languages with E5 embeddings, ChromaDB vector store, Gemini 2.5 Flash LLM, and NLLB-200 translation. Ask questions in any language, get accurate answers with citations

本地化医学文献 RAG 问答系统。覆盖毕业论文文献调研、科研综述撰写、临床指南查阅、药学信息查询等场景 ，同时具备小数据量下的生产级可用性。支持多篇中英文医学论文 PDF（糖尿病、心血管、BMI 等）的智能检索与生成式问答。基于 Faiss 语义向量 + BM25 关键词 + Cross-Encoder 重排序三重检索 pipeline，模块化设计可自由切换嵌入模型、生成模型与重排序模型。全程不依赖任何第三方 API，基于 Ollama 本地部署 bge-m3 嵌入模型与 qwen2:7b 生成模型，确保医学数据隐私安全。实现了多会话隔离、流式 SSE 输出、异步 PDF 上传、前端历史持久化等生产级特性。

支持多模型对话、PDF知识库问答的AI应用，自带暗黑模式，本地一键部署

Sleek Streamlit chat app for Google Gemini (Files API). Dark, gradient UI with model picker, usage dialog, file/image/audio/PDF attach & preview, chat history, image persistence, robust error handling, and token usage tracking. Supports streaming replies and modular backend via google-genai.

AI-powered PDF Q&A chatbot. Upload any document and have a real conversation with it. Built with RAG architecture using LangChain, Groq (Llama 3.3-70B), ChromaDB, and HuggingFace embeddings, completely free to run.

Streamlit-based chatbot to interact with PDFs using Retrieval-Augmented Generation (RAG), FAISS, Sentence Transformers, and Mistral LLM

A local PDF chatbot that uses RAG to answer questions from a single text-based PDF with citations and out-of-scope detection.

Document Q&A system using RAG, FAISS vector search and LLM inference

Enables context-aware question answering over PDFs using retrieval-augmented generation with vector embeddings. Built with Next.js App Router and OpenAI models for low-latency document search and response generation.

Local PDF Q&A with Ollama

Web app to query PDFs in natural language. Upload a PDF, ask questions, get answers with source pages. RAG (Chroma + BM25) + OpenAI; Vue 3 frontend.

A high-performance Speculative RAG pipeline designed to reduce latency by combining fast draft generation and accurate verification using Groq Llama models, local HuggingFace embeddings, ChromaDB vector search, and end-to-end observability with Langfuse.

RAG-based PDF Q&A system. Upload any PDF, ask questions, get answers grounded in the document. Built with LangChain, FAISS, BGE embeddings, Groq (LLaMA 3.3 70B), and Streamlit.

Evidence-grounded PDF question answering with multi-layer hallucination defense. Refuses to answer when evidence is insufficient. Fully local — runs on Ollama + ChromaDB + Streamlit. No cloud APIs.

Streamlit PDF question-answering app using LangChain, FAISS retrieval, and Groq chat models