<aside> 💡

相關頁面: AI相關工具非常非常的多，RAG開發工具會整理在RAG開發工具，LLM以及相關工具會整理在Large Language Models，Graph RAG開發工具會整理在Graph RAG，其他工具整理在Generative AI開發工具中。目前還有一些資料在Generative AI及RAG中，將逐步整理這些內容。

</aside>

RAG開發工具簡介

RAG開發工具基本上跟Gen AI、Agent有滿高的重複性，現在Agent越來越紅了，所有工具都不再強調自己是RAG開發工具了。比較不一樣的是跟Retrieval有關的工具，例如Embedding、Vector Database。

Web Framework for RAG

• 開源RAG網頁套件研究
  • open-webui
  • Hollama
  • LoLLMs Webui
  • BionicGPT
  • LibreChat
  • LLMStack

Embeddings

Vector Databases

資料

PageIndex

• RAG is Not Dead! No Chunking, No Vectors, Just Vectorless to Get the Higher Accuracy (Friend Link)
  • the PageIndex method converts a document into a hierarchical tree structure (similar to a table of contents), and LLM searches through that structure, making it possible to understand the context and find the information you need, just like a human reading a document.

Bright Data

• How to Build a Production-Ready RAG App with Gemma and Bright Data in Under an Hour
  • Gemma 3 via Ollama
  • ChromaDB for embeddings + search
  • Bright Data’s SERP for live context injection

RAG Frameworks

多數的框架適用於RAG也適用於其他Gen AI的情境(如: AI Agent)，更多內容請參考Generative AI 開發工具 。

• RAG Frameworks Explored: LlamaIndex vs. LangChain for Next-Gen LLMs (Friend Link)
  • LlamaIndex is more beginner-friendly due to its simple and intuitive implementation, whereas LangChain requires a deeper understanding of LLMs and NLP concepts.
• Top 10 RAG Frameworks Github Repos 2024
  • Haystack by deepset-ai
  • RAGFlow by infiniflow
  • txtai by neuml
  • STORM by stanford-oval
  • LLM-App by pathwaycom
  • Cognita by truefoundry
  • R2R by SciPhi-AI
  • Neurite by satellitecomponent
  • FlashRAG by RUC-NLPIR
  • Canopy by pinecone-io
• LLM Frameworks in Action: Building RAG Systems with LangChain, LlamaIndex, and Haystack!
  • Vectorize: Building RAG Pipelines Made Easy!

Azure

• Build Intelligent RAG with Image understanding & hierarchical document structure analysis with Azure
• RAG & AI Agent using Azure AI Stack