PrivyPDF

Chat with PDFs using your own local AI. Version 0.4.1 adds a cleaner iOS layout, smoother PDF chat, better settings, and improved LLM setup.

PrivyPDF lets you chat with your PDF documents using your own AI—powered by Ollama on your local network. Your data stays on your device and your LAN.

FEATURES • Chat with the current PDF: ask questions and get answers grounded in the document • Local AI via Ollama: set your server URL and choose Chat, Embedding, and OCR models in Settings • Parse & Index: extract text and optional embeddings; trigger automatically on import (when Ollama is configured) or manually from Library or Chat • Remove documents: long-press a PDF in Library to remove it • Export answers to Markdown, JSON, or Word (RTF) • Library: import, open, and manage multiple PDFs • First-time setup: prompt to configure Ollama Base URL and models

SETTINGS • Ollama Base URL, request timeout • Chat, Embedding, and OCR model pickers (refresh from Ollama) • Model temperature slider • Use Embeddings toggle for semantic retrieval • Retrieval options: TopK, history turns, chunk size

Requires an Ollama server on your network (e.g. http://192.168.x.x:11434). or LM Studio on your network (e.g. http://192.168.x.x:1234/v1) No cloud; all processing is local or on your LAN.