timmyy123/LLM-Hub

Local LLM, image&video&music generator, vibecode like cursor with local models on your phone

What it solves

LLM Hub is a privacy-focused mobile application that allows users to run large language models (LLMs), image generation, and video generation entirely on-device. It eliminates the need for cloud-based AI services, ensuring that conversations and data never leave the device while providing offline access to AI tools.

How it works

The app leverages mobile-specific hardware acceleration (CPU, GPU, and NPU) to run local inference. On Android, it uses the GenieX SDK and WhisperKit for ASR; on iOS, it utilizes the Run Anywhere SDK, MediaGenerationKit, and whisper.cpp. It supports a wide variety of model formats including .gguf, .litertlm, and .task, and allows users to import custom models or download them directly from HuggingFace.

Who it’s for

Users who prioritize data privacy and offline AI capabilities on their smartphones (Android and iOS) and developers who want to integrate on-device AI runtimes into mobile apps.

Highlights

  • Multi-modal AI Suite: Includes local chat with RAG memory, image generation (Stable Diffusion), video generation (Stable Video Diffusion), and on-device transcription (Whisper).
  • Autonomous AI Agent: An Android-specific feature that can execute Termux shell commands, manage device tools (calendar, alarms, flashlight), and perform self-diagnostics to auto-correct command errors.
  • Vibe Coder: A real-time app builder that generates code and provides a live HTML/JS preview.
  • UTF-8/NPU Acceleration: Optimized for high performance using mobile NPUs for tasks like image upscaling (RealESRGAN) and speech-to-text.
  • Privacy-First: 100% on-device processing with zero data collection and no account requirements.

Related

  • Project
    dineshsoudagar/local-llms-on-android

    An Android application that enables fully offline, private LLM chat with support for voice, image, and camera input using ONNX and LiteRT backends.

  • Project
    google-ai-edge/gallery

    A mobile and macOS application for running open-source LLMs locally on-device, providing a private and offline environment to experience and evaluate generative AI.

  • Project
    oobabooga/textgen

    A private, open-source desktop app for running local LLMs with support for multiple backends, multimodal input, tool-calling, and local fine-tuning.

  • Project
    off-grid-ai/OGAM

    A comprehensive offline AI suite for mobile and Mac that provides text, image, and vision AI, and voice transcription, all running natively on-device for total privacy.

  • Project
    Ryan-yang125/ChatLLM-Web

    A private, browser-native AI studio and agent workspace that runs LLMs locally on your device using WebGPU, requiring no API keys or backend servers.