GitHub - psychic-api/rag-stack: 🤖 Deploy a private ChatGPT alternative hosted within your VPC. 🔮 Connect it to your organization's knowledge base and use it as a corporate oracle. Supports open-source LLMs like Llama 2, Falcon, and GPT4All.
🔮 Enhance your VPC with Psychic API's RAGstack! 🤖 Deploy a private ChatGPT alternative integrated with LLMs like Llama 2 and GPT4All. 🌟 Access an oracle for your organization's knowledge base! #AI #RAGstack #LLMs #GPT4All 🚀
- The Psychic API's RAGstack allows the deployment of a private ChatGPT alternative within a VPC, enhancing it with data retrieval capabilities.
- RAGstack integrates various open-source LLMs such as Llama 2, Falcon, and GPT4All for augmented generation.
- Retrieval Augmented Generation (RAG) improves LLMs by fetching external information via prompts, enhancing their understanding beyond training data.
- RAGstack resources include GPT4All, Falcon-7b on GPU, Llama 2 on GPU, and the Qdrant vector database, catering to different deployment scenarios.
- The setup process involves running scripts for local deployment, Google Cloud deployment, AWS deployment, and Azure deployment.
- Terraform is used for implementing deployment scripts, ensuring flexibility and scalability in deploying the RAGstack.
- Notable features include GPT4all and Falcon-7b support, multiple deployment options, and ongoing development for Llama-2-40b support.