How to increase the file upload size limit in OP Vault?

Edit the MAX_FILE_SIZE and MAX_TOTAL_UPLOAD_SIZE constants in the fileupload.go file, as explained in the 'Uploading larger files' section of the README, though this requires code changes and server restart.

Can OP Vault work with local LLMs instead of OpenAI?

No, it's tightly integrated with the OpenAI API for embeddings and prompting; switching to local LLMs would require significant modifications to the Golang backend and embedding logic.

OP Vault vs LangChain for document question-answering?

OP Vault is a ready-to-use application with a frontend, ideal for quick deployment, while LangChain is a framework offering more flexibility but requiring custom development for similar functionality.

How to deploy OP Vault on a cloud server?

The README doesn't provide detailed deployment guides, but since it uses Golang and React, you'll need to build the backend, serve the frontend, and set up environment variables for API keys, similar to typical web app deployments.

Does OP Vault support image files or OCR for scanned documents?

No, it only supports text-based formats like PDF, txt, epub, and .docx; image files or OCR capabilities are not included, limiting use with non-textual content.

What are the costs of running OP Vault long-term?

Costs include OpenAI API usage for embeddings and queries, plus Pinecone storage fees, which can scale with document volume, making it expensive for large knowledge bases.

Open-Awesome

OP Vault ChatGPT

MITJavaScript

Give ChatGPT long-term memory by uploading custom knowledge base files (PDF, txt, epub) and asking questions via a React frontend.

Visit Website GitHub

3.4k stars297 forks0 contributors

What is OP Vault ChatGPT?

OP Vault is an open-source tool that extends ChatGPT with long-term memory by allowing users to upload custom knowledge base files like PDFs, txt, and epub documents. It uses the OP Stack (OpenAI + Pinecone Vector Database) to process, store, and retrieve information from these documents, enabling precise question-answering based on the uploaded content. The project solves the problem of AI models lacking access to private or specific knowledge sources.

Target Audience

Developers, researchers, and organizations looking to build custom AI-powered knowledge bases or enhance ChatGPT with domain-specific information from documents.

Value Proposition

It provides a self-hostable, user-friendly interface for uploading and querying documents, with accurate answers and source citations, making it a practical alternative to manual document search or proprietary knowledge management tools.

Overview

OP Vault ChatGPT: Give ChatGPT long-term memory using the OP Stack (OpenAI + Pinecone Vector Database). Upload your own custom knowledge base files (PDF, txt, epub, etc) using a simple React frontend.

Use Cases

Best For

Building a personal AI assistant with access to private documents
Creating a searchable knowledge base from PDFs and ebooks
Adding long-term memory to ChatGPT for specific domains
Extracting insights from large collections of text documents
Self-hosting a document Q&A system with source attribution
Experimenting with OpenAI and Pinecone vector databases

Not Ideal For

Projects requiring fully offline or local document search without external API dependencies
Teams with limited DevOps resources for maintaining a multi-language stack (Golang, React, node)
Applications needing real-time ingestion and querying from streaming data sources
Organizations with strict data privacy rules prohibiting cloud-based AI services like OpenAI and Pinecone

Pros & Cons

Pros

Multi-Format Document Support

Supports PDF, txt, epub, .docx, and other formats via the React frontend, allowing diverse knowledge base creation from common file types.

Accurate Source Attribution

Provides answers with citations to specific filenames and context snippets, as shown in the screenshots, ensuring traceability to original content.

Scalable Vector-Based Retrieval

Leverages Pinecone for efficient storage and retrieval of embeddings, enabling handling of entire libraries of documents, as highlighted in the README.

Self-Hostable Deployment

Offers a Golang backend and React frontend for full control over deployment, making it suitable for custom or private installations.

Cons

Complex Dependency Setup

Requires manual installation of specific versions like node v19, go, and poppler, with no containerization or one-click setup, increasing initial effort.

Reliance on Paid External APIs

Depends on OpenAI and Pinecone APIs, incurring ongoing costs and subject to free-tier restrictions, such as Pinecone's namespace limits mentioned in the README.

Limited Out-of-the-Box Features

Lacks built-in user authentication, access controls, or advanced querying options, focusing only on basic upload and Q&A without extensibility hooks.

Frequently Asked Questions

Related Projects

ChatGPT Next Web

Stars88,542

Forks59,381

Last commit18 days ago

Open Interpreter

A coding agent for open models like Kimi K3

Stars67,209

Forks5,769

Last commit5 days ago

FastChat

An open platform for training, serving, and evaluating large language models. Release repo for Vicuna and Chatbot Arena.

Stars39,501

Forks4,786

Last commit2 months ago

AgentGPT

🤖 Assemble, configure, and deploy autonomous AI Agents in your browser.

Stars36,287

Forks9,297

Last commit1 year ago

Community-curated · Updated weekly · 100% open source

Found a gem we're missing?

Open-Awesome is built by the community, for the community. Submit a project, suggest an awesome list, or help improve the catalog on GitHub.

Submit a project Star on GitHub