OpenCone

On-device Retrieval Augmented Generation for iOS. Extracts text from PDFs and images via Vision OCR, generates embeddings, syncs with Pinecone, and retrieves grounded answers.

On-Device RAG Client

A native SwiftUI client for document extraction and Pinecone vector database integration.

OpenCone is a native iOS application that performs local document ingestion (PDF, DOCX, TXT, and images via OCR), generates vector embeddings via OpenAI, syncs with Pinecone vector search, and runs Retrieval Augmented Generation (RAG).

Key Features of OpenCone

Multi-Format Document Management

Upload, view, and manage various document types (PDF, DOCX, TXT, images with OCR). Securely handles file access using bookmarks.

Advanced Processing Pipeline

Automated text extraction, strategic chunking (MIME-type specific), and robust error handling during document ingestion.

OpenAI Embeddings

Generates high-dimensional vector embeddings for text chunks using configurable OpenAI models (e.g., text-embedding-3-large).

Pinecone Integration

Full lifecycle management for Pinecone serverless indexes: create, list, select active index, and manage namespaces.

Semantic Search & RAG

Perform natural language queries. Retrieves semantically similar text chunks and generates AI answers via OpenAI.

Detailed Statistics

View comprehensive processing statistics including phase timings, token counts, and chunk size distributions.

Custom Design & Theming

Features a bespoke UI library (OCDesignSystem) with OCButton, OCCard, OCBadge. Multiple themes available.

Real-time Processing Logs

Detailed, filterable logs for all major operations with timestamps and context.

Guided Setup

User-friendly WelcomeView for initial API key configuration and app introduction.

How OpenCone Works

1

Launch & Setup

Enter your OpenAI API Key, Pinecone API Key, and Pinecone Project ID through the guided WelcomeView.

2

Add & Process Documents

Upload files (PDF, TXT, DOCX, images), create or select a Pinecone index, then process documents to extract text, generate embeddings, and upload to Pinecone.

3

Semantic Search

Ask natural language questions. The app retrieves semantically similar chunks and generates AI-powered answers using RAG.

4

Monitor & Configure

View real-time logs, check processing statistics, and adjust settings including models, chunk sizes, and themes.

Tech Stack & Architecture

Architecture

  • Pattern: MVVM with Service Layer
  • UI: SwiftUI
  • Concurrency: async/await & Combine
  • Design System: Custom OCDesignSystem

Integrations

  • OpenAI: Embeddings & Completions
  • Pinecone: Vector Database
  • PDFKit: PDF Processing
  • VisionKit: OCR for Images

App Store Connect Metadata