Use private LLMs fully offline with on-device model processing for Apple silicon. Chat with or generate outputs for other apps. Customize theme, fonts, and system prompts. Works on iOS, iPadOS, macOS, and visionOS. Open source and free.
Run AI models like Llama, Gemma, and Qwen offline on iPhone and iPad. Access features without internet. Store data with no login or data collection. Use Apple MLX for optimized performance on Apple Silicon devices.
Manage Wi-Fi settings by viewing, updating, and sharing the name and password. Monitor connected devices, rename them, and add to profiles. Set downtime schedules to control screen time for family members.
Access and interact with private local AI models or open-source language models via OpenRouter API. Chat with customizable models, run smaller LLMs securely on-device, and connect to your own hosted models.
Remote server access allows connection to self-hosted servers from any location. Per- system prompts let users customize conversation contexts. Prompt editing and regeneration enable refining interactions.
Allows users to interact with AI models through a chat interface. It supports multiple chat modes including Simple and Agent. The runs on-device AI inference using .cpp optimized for Metal. It includes support for Meta's and Google's Gemma models.
Learn to sketch figures and garments with step-by-step guidance. Create and manage layers for detailed illustrations. Use AI tools to generate clothing and accessories in specific styles. Add body types including models, women, men, and plus size.
Text in real time across multiple platforms. Supports slang and internet culture. Integrates with social media and messaging apps. Offers context-aware translations. Allows users to switch languages instantly.