LLM Hub

Experience the next generation of AI with LLMHub, the most powerful private AI assistant for iPhone and iPad. Run state-of-the-art Large Language Models (LLMs) completely on-device with zero data sharing and no internet required. Powered by the RunAnywhere SDK and optimized for Apple Silicon.

EIGHT POWERFUL ON-DEVICE TOOLS SMART CHAT Multi-turn conversations with context awareness, RAG memory, and optional text-to-speech auto-readout. Supports Text, Images, and Audio inputs. Powered by the RunAnywhere inference engine for maximum performance.
WRITING AID Summarize documents, rewrite content, and improve grammar instantly. Process long-form text locally without ever sending a single word to the cloud.

IMAGE GENERATOR Create stunning visuals with Stable Diffusion. Optimized for the Apple Neural Engine (ANE) and Metal for fast, local image synthesis directly from text or image prompts.

TRANSLATOR (50+ Languages) Real-time, bidirectional translation for text, audio, and images (OCR). Perfect for secure, private communication while traveling or working.

TRANSCRIBER High-accuracy speech-to-text conversion using native iOS recognition. All audio processing happens on-device for total privacy.

VIBECODE CREATOR The ultimate AI personalization engine. Build, share, and use custom AI personas with deep-level behavioral instructions that persist across all your chats.

CODE ASSISTANT Generate, explain, and debug code using VibeCode-enhanced models like Gemma-4. Features local reasoning that understands your specific programming style.

SCAM DETECTOR Analyze suspicious messages and phishing images for risk. Get clear local assessments to keep your digital identity safe.

CUTTING-EDGE TECHNOLOGY STACK
LLMHub is built specifically for iOS, leveraging the latest in mobile AI research:

• RunAnywhere SDK: A high-performance inference backend supporting GGUF, ONNX, and LiteRT formats with full Metal and Accelerate acceleration. • Apple Foundation Model Support: Seamlessly integrate with native Apple Intelligence foundation models for high-speed, system-level efficiency. • Metal & Neural Engine Optimization: Desktop-class performance on your iPhone via Apple Silicon.

FEATURED ON-DEVICE MODELS
• Gemma-4 (Google) - The latest in multimodal performance, featuring E2B and E4B variants for advanced text and vision tasks. • Apple Foundation Model - Apple's system-integrated intelligence optimized for your specific hardware. • Llama-3.2 (Meta) - Powerful 1B and 3B reasoning models. • Phi-4 Mini (Microsoft) - Optimized for mobile productivity. • Ministral (Mistral AI) - High-performance multilingual support.

COMPLETE PRIVACY GUARANTEED
• No Internet Required: All AI inference happens 100% on your device. • No Data Sharing: We never collect, store, or see your data. • No Accounts: Start using advanced AI immediately with no tracking or cloud uploads. • Open Source: Transparent and secure (MIT License).

ADVANCED IOS FEATURES
• Multimodal Intelligence: Interact via text, voice, or photos. • Direct HuggingFace Downloads: Import any compatible GGUF or ONNX model directly to your personal library. • Native SwiftUI Interface: Beautifully designed for iOS with a focus on fluidity and speed. • Thinking Models: Experience next-gen reasoning with models that "think" before they speak.