A curated list of tools, frameworks, and resources for Retrieval-Augmented Generation (RAG) — systems that combine LLMs with external knowledge retrieval for more accurate, grounded answers.
- Frameworks
- Vector Databases
- Embedding Models
- Pipelines & Orchestration
- Evaluation & Benchmarking
- Examples
- Learning Resources
- Related Awesome Lists
- LlamaIndex – Data framework for building RAG apps with retrieval, agents, and advanced pipelines.
- LangChain – Industry-standard framework with RAG chains, retrievers, memory, and eval tools.
- Haystack – End-to-end RAG pipeline framework with search, indexing, and orchestration.
- Marqo – Neural search engine for RAG with built-in vector indexing and document enrichment.
- OpenSearch – Search engine with vector search and hybrid retrieval for RAG setups.
- Jina – Framework for multimodal retrieval and generative search.
- Milvus Lite – Lightweight local vector DB ideal for prototyping RAG apps.
- Pinecone – Managed vector database optimized for RAG at scale.
- Weaviate – Open-source vector DB with hybrid search, modules, and semantic tooling.
- Milvus – High-performance, cloud-native vector DB.
- qdrant – Open-source vector search engine with filters, HNSW, and payloads.
- Chroma – Local-first vector database widely used in prototyping and small deployments.
- Redis Vector Search – Redis module for hybrid text + vector search.
- OpenAI Text Embeddings – High-quality embeddings for semantic search and RAG.
- HuggingFace Embeddings – Library of open embedding models from small to large.
- Voyage Embeddings – High-performance multilingual and domain-specific embeddings.
- Cohere Embeddings – Strong semantic embeddings with reranking support.
- Nomic Embeddings – Embeddings designed for large-scale, interactive datasets.
- LangGraph – Graph-based RAG workflows with state management and multi-step reasoning.
- LlamaIndex Pipelines – End-to-end pipelines for chunking, retrieval, reranking, and evaluation.
- Deepset Cloud – Fully managed RAG pipelines built on Haystack.
- FastRAG – Minimal, efficient RAG pipeline with reranking.
- RAGAS – Automated evaluation metrics for RAG quality, grounding, and hallucinations.
- TruLens – Evaluation and monitoring for RAG pipelines.
- Arize Phoenix – Open-source LLM observability and RAG evaluation toolkit.
- LangChain Evaluators – Built-in evaluators for RAG, retrieval, and chain performance.
- LlamaIndex RAG Examples – End-to-end tutorials and sample apps.
- LangChain RAG Templates – Official RAG app templates including agents + retrieval.
- Haystack RAG Examples – Starter implementations for RAG pipelines.
- Multimodal RAG (Jina) – Image + text RAG examples using Jina.
- OpenAI RAG Cookbook Examples – Retrieval + function calling + tool use demos.
- RAG 101 – LangChain – Intro guides for building RAG pipelines.
- LlamaIndex RAG Guide – High-level and deep-dive RAG documentation.
- Haystack RAG Theory – Concepts, architectures, and tutorials.
- Multimodal RAG Guides – Tutorials for text, image, and video retrieval.
- Advanced RAG: Retrieval, Reranking, and Tool Use – Topic-level overview across GitHub.
Contributions are welcome. Please ensure your submission fully follows the requirements outlined in CONTRIBUTING.md, including formatting, scope alignment, and category placement.
Pull requests that do not adhere to the contribution guidelines may be closed.
