Process Flow
Overview
Generative AI is a game changer for our society, but adoption in companies of all sizes and data-sensitive domains such as legal is limited by a clear concern: privacy. Not being able to ensure that your data is fully under your control when using third-party AI tools is a risk this industry cannot take.
Personal GPT was born out of a simple yet profound realization: the need for powerful AI that respects and protects your confidentiality. Founded by the innovative minds behind PromptMule and brought to life in collaboration with Numino, its mission was to revolutionize the way users interact with AI while safeguarding their sensitive information.
This project was a prototype developed in collaboration with a few private clients who share our vision for data privacy and security. To ensure robust and scalable performance, Private GPT is hosted using the power of AWS Cloud. This partnership allowed us to provide reliable, secure, and high-performance AI services that our clients could trust.
PersonalGpt is a production-ready AI project where
- Users can ask questions about their documents using the power of Large Language Models (LLMs), even in scenarios without an Internet connection.
- 100% private, no data leaves the user's execution environment at any point.
The application offers all the primitives required to build private, context-aware AI applications
Category
Cloud Backend, AI
Client
Promptmule
Tags
- AWS
- LLama2
- FAISS
- Langchain
- Chainlit
Product Features
Ingestion of documents
internally managing document parsing, splitting, metadata extraction, embedding generation and storage.
Chat & Completions using context
abstracting the retrieval of context, the prompt engineering and the response generation.
Technology Stack
Architecture
LLaMA 2 model
Is pre-trained and fine-tuned with 2 Trillion tokens and 7 to 70 Billion parameters which makes it one of the powerful open source models. It outperforms other open source LLMs on many external benchmarks, including reasoning, coding, proficiency, and knowledge tests.
LangChain
A powerful framework specifically designed for developing language model-driven applications, serves as the foundation for our project. It provides us with the necessary tools and capabilities to create an intelligent system that can accurately answer questions based on specific documents.
FAISS (vector data store)
To enhance the performance and efficiency of our question-answering system, we integrate FAISS, an efficient vector database known for building high-performance vector search applications. By leveraging its capabilities, we can significantly improve the speed and accuracy of our system’s search and retrieval processes.
Chainlit
An open-source Python package that makes it incredibly fast to build Chat GPT like applications with your own business logic and data. It enables the user to have an interactive discussion with the model.
Product Screenshots
Conclusion
While fine-tuned GPT models might generate answers based on the general knowledge embedded in the model, which might be less precise or unrelated to the question’s context, PersonalGpt can generate more context-specific and precise answers by grounding answers in specific passages from relevant documents.
By harnessing the capabilities of semantic search and large language models & AWS cloud services, this approach offers a robust and versatile solution for extracting valuable information from extensive document collections while ensuring that the confidential information is kept secure within the users premise.
