UltraRAG v2: A Low-Code MCP Framework for Building Complex and Innovative RAG Pipelines

A demonstration of hybrid search with reranking using Qdrant and BGE-M3 model. A showcase of dense and sparse retrieval combined with ColBERT reranking for optimal search results

ONNX implementation of the BGE-M3 multilingual embedding model and tokenizer with native C#, Java, and Python implementations. Generates all three embedding types: dense, sparse, and ColBERT vectors.

LLM API Server , OpenAI 同时支持 ChatGLM3 ，Llama, Llama-3, Firefunction, Openfunctions ，BAAI/bge-m3 ,bge-large-zh-v1.5

Example application for using the BGE-M3 embedding model and Google's Gemma-2-9B-Instruct generation model in a LangChain-based RAG pipeline to answer Lord of the Rings trivia questions

Asistente RAG que ingiere la carpeta de proyectos local, construye un índice en ChromaDB con embeddings avanzados BGE‑M3 y responde preguntas a través de un LLM DeepSeek.

Chinese Fake News Detection based on MacBERT

🚀 企业级私有化 RAG 知识库系统 | 💯 纯离线部署、无需联网 | 🔐 支持 LDAP/AD 统一认证与 RBAC 部门级权限隔离 | ⚡️ 基于 LlamaIndex + Ollama + BGE-M3/Rerank | 📝 支持多格式文档解析、上下文对话记忆及精准的 PDF 原文溯源高亮 | 📊 内置数据可视化仪表盘（🚀 Enterprise RAG Knowledge Base. 💯 Fully offline/air-gapped deployment. 🔐 Features LDAP/AD integration, RBAC & Department Isolation. ⚡️ Powered by LlamaIndex, Ollama, and Local R

An intelligent Retrieval-Augmented Generation (RAG) system that helps students find answers within video lectures using semantic search and LLMs (Ollama/GPT). Automatically extracts subtitles, generates embeddings, retrieves relevant clips, and provides context-aware answers with timestamps.

Hybrid RAG for Thai ROV patch notes: normalize diffs → hybrid retrieve (BM25 + dense in Chroma, MMR, optional bge reranker) → citation-first answers via FastAPI with OpenAI/Ollama generators. | ระบบ RAG ภาษาไทยสำหรับแพตช์โน้ต ROV

Herramientas para procesar datos de desaparecidos: anonimización de narrativas con IA. Búsqueda semántica de casos similares mediante embeddings de texto (BGE-M3).

Train a pooling Head for Dense Embeddings

Telegram chatbot FAQ untuk membantu mahasiswa Universitas Lampung mengakses informasi Siakadu menggunakan semantic search (BGE-M3 + FAISS).

Multimodal Answer Engine generates answers with images and text. Uses only open source models with vllm inferencing with eagle speculative decoding.