Train Models Contrastively in Pytorch

Radient turns many data types (not just text) into vectors for similarity search, RAG, regression analysis, and more.

Think-on-Graph 3.0: Efficient and Adaptive LLM Reasoning on Heterogeneous Graphs via Multi-Agent Dual-Evolving Context Retrieval

Production inference for encoder models - ColBERT, GLiNER, ColPali, embeddings etc. - as vLLM plugins for online and in-process deployment

A sample app for the Multimodal Retrieval-Augmented Generation pattern running in Azure, using Azure AI Search for retrieval and Azure OpenAI large language models to power Q&A experiences.

Awesome Memory Papers in Vision-Language Models

High-performance late-interaction retrieval engine for on-prem AI. ColBERT/ColPali multi-vector search with Rust fused MaxSim, Triton GPU kernels, ROQ quantization, LEMUR routing, WAL-backed CRUD, and a FastAPI server — single machine, CPU or GPU.

Build sovereign RAG systems with MAS‑RAG, Dual‑RAG, GraphRAG, Spatial‑RAG, multimodal pipelines, and vector search directly inside Oracle AI Database 26ai and Exadata.

🧠 Multimodal Retrieval-Augmented Generation that "weaves" together text and images seamlessly. 🪡

[NAACL 2024] Official Implementation of paper "Self-Adaptive Sampling for Efficient Video Question Answering on Image--Text Models"

🔰 A Comprehensive RAG repository covering basic vanilla RAG techniques, advanced retrieval methods, hybrid search fusion approaches, hands-on reranking techniques with code + explanation 📚✨

🚀 HAG: Next-Gen AI | Neo4j + Weaviate Fusion | Dual-Similarity Retrieval | 100% Local & Private | Graph Intelligence Meets Vector Search

OpenAI-compatible multimodal embedding server for Qwen3-VL-Embedding-2B — embed text, images, or both via a simple REST API.

[EMNLP 2025] M-LongDoc: A Benchmark For Multimodal Super-Long Document Understanding And A Retrieval-Aware Tuning Framework

Multimodal RAG Production

Self-adaptive Planning Agent。自适应规划代理的多模态检索增强生成技术。

30 production-focused examples: covering RAG patterns, hybrid search, routing, agentic RAG, knowledge graphs, multimodal RAG, and debugging.

Anaya is a Content Engine that specializes in analyzing and comparing multiple PDF documents. It uses Retrieval Augmented Generation (RAG) techniques to effectively retrieve, assess, and generate insights from the documents.

A doctor-assistive AI system that interprets medical knowledge and patient images simultaneously. It utilizes a Dual-Encoder architecture to cross-reference textbook theory with visual pathology, generating clinically grounded diagnoses.

📄 Multimodal RAG pipeline combining ColPALI visual retrieval, YOLO-DocLayNet layout detection, sentence embedding-based text retrieval, and LLaMA-4 completion for document question answering.