VecML vs LLM Pigeon

VecML APP offers tools for intelligent photo search and large language models (LLMs) fully on device. Users can select from a range of small LLMs and engage in conversations with documents stored locally. The app can process files (PDFs, DOC documents, images, etc.) to generate embedding vectors, which are stored locally on the device. These embedding vectors are crucial for RAG (retrieval-augmented generation) capabilities, enhancing the LLM's ability to provide more accurate and relevant responses when queries are related to the local files.

Put a full-featured AI assistant in your pocket—without sending a single word to any outside cloud. LLM Pigeon works together with LLM Pigeon Server on your Mac. • Type a message on your iPhone. • The text is stored inside your private iCloud files. • Your Mac picks it up, runs the large-language model locally, writes the reply, and CloudKit delivers the answer back to the phone. Why you’ll love it • 100 % privacy: messages never leave your own iCloud container; no analytics, no ads, no accounts. You can optionally add a search-engine API. • Real LLMs: choose from a set of built-in models or point the server at LM Studio or Ollama to swap in any model you prefer. How it works 1. Install LLM Pigeon Server (free) on a Mac that’s signed in with the same iCloud account. 2. Download the models you want and, if you’d like web search, enter a Serper API key. 3. Start a new chat in the iOS app. 4. Receive answers in seconds, even on hotel Wi-Fi. Zero data collection • No personal information • No usage analytics • No crash logs—everything stays on your devices • If you enable web search, only the queries your LLM generates are sent to Serper. Requirements • iPhone running iOS 17 or later • Mac running macOS 14 (Sonoma) or later with LLM Pigeon Server installed • iCloud Drive enabled on both devices