Honey LLM — Offline AI
Practical generative AI designed to run locally on mobile hardware without requiring continuous cloud connectivity or third-party data transmission.
Honey LLM on Android
Official Google Play listing imagery. Features shown are Android-specific and may change between releases. Scroll to explore the screenshots.
Honey LLM on iPhone
Official App Store listing imagery. The iPhone feature set differs from Android; scroll to explore the screenshots.
Privacy by Architecture, Not Policy
Most AI assistants depend on transmitting your queries, conversations, and personal documents to remote server clusters for processing. Honey LLM operates entirely on the device: once models are downloaded, all inference, token generation, and context evaluation occur locally in memory.
Available now: Honey LLM has public listings on Google Play for Android and the Apple App Store for iPhone/iPad. Features and supported hardware vary between platforms.
The Android edition offers private document assistance and an encrypted Vault. The iPhone edition offers local offline chat, model downloads, and model switching; do not assume feature parity between platforms.
Core Capabilities
- Native On-Device Inference: Runs quantized language models locally using a native llama.cpp integration on supported Android devices.
- Local Model Management: Select, download, inspect, and remove models locally without server dependencies.
- Private Document Assistance: Evaluate personal notes and documents in private storage without cloud synchronization.
- Completely Offline Operation: Full conversational functionality remains available on flights, remote job sites, and private environments.
Technical Requirements & Compatibility
Running large language models locally requires capable mobile hardware. Recommended specifications:
- Operating System: Android 8.0 (API 26) or newer; device compatibility and performance vary.
- Memory (RAM): Requirements vary by selected model and available device memory. Smaller models may be better suited to lower-memory devices.
- Storage: Sufficient free space for the selected downloadable model and application data; model sizes vary.
- Hardware Support: 64-bit ARM (arm64-v8a).
iPhone and iPad
The iOS version has its own requirements and feature set. The public App Store listing specifies iOS/iPadOS 18 or later. Check the App Store compatibility section for your device before installing.
Support & Further Questions
Have questions about model support, performance, or recovery? Visit our dedicated support resources: