Awesome GitHub RepositoriesLocal On-Device AI

Chat2DB is an AI-powered SQL client and multi-database GUI manager designed for managing various relational and NoSQL database systems. It serves as a visual database management tool and a natural language to SQL interface, allowing users to convert plain text descriptions into executable and optimized queries. The platform distinguishes itself through automated business intelligence capabilities, which include the generation of real-time data visualization dashboards and AI-driven data analysis from spreadsheets. To ensure data privacy, it supports secure local AI deployment, enabling large

MoneyPrinter is an automated short-form video creation pipeline that generates complete YouTube Shorts from a given topic. It combines local LLM-powered script generation with programmatic video assembly, all managed through a database-backed job queue for reliable, restart-tolerant processing. The system uses an Ollama-powered local language model to write video scripts and metadata entirely on-device, keeping data private and offline. It then produces the final video clip using MoviePy for compositing clips, text, and audio, creating a complete YouTube Shorts video without manual editing. V

The simplest way to run LLaMA on your local machine

This project is a local AI inpainting tool designed to erase hard-coded subtitles and watermarks from videos and images. It functions as a content-aware media restorer that uses deep learning to reconstruct missing pixels and preserve the original resolution of the source files. The software is distinguished by its local execution model, running inference on host hardware to process media without relying on external cloud APIs. It employs content-aware model selection, allowing the use of different generative algorithms based on media types, such as animation or live action, to optimize visua

WasmEdge is an extensible WebAssembly runtime that executes WebAssembly bytecode in a secure sandbox for cloud, edge, and embedded applications. It functions as a multi-language compiler, compiling applications written in Rust, JavaScript, Go, and Python into WebAssembly bytecode for sandboxed execution, and as a server-side JavaScript runtime that runs JavaScript programs with ES6 modules, NPM packages, and Node.js-compatible APIs. The runtime also serves as an AI inference runtime, executing AI models from JavaScript using WASI-NN plug-ins for inference tasks on personal devices and edge har

Local Deep Researcher is a fully local web research assistant that uses any LLM hosted by Ollama or LMStudio. Give it a topic and it will generate a web search query, gather web search results, summarize the results of web search, reflect on the summary to examine knowledge gaps, generate a new…

Reor is a local AI knowledge management application that stores, links, and searches personal notes using large language models and vector embeddings entirely on the user's device. It functions as a private AI note assistant, keeping all data and processing local for full privacy without relying on external cloud services. The application integrates with Ollama to manage the lifecycle of local LLMs and embedding models, handling downloads, updates, and execution. Notes are imported from markdown files, preserving existing file structure, and are automatically linked through vector-similarity

Page Assist is a browser-based AI integration tool that provides a sidebar interface for interacting with AI models while browsing the web. It focuses on privacy-focused chatting and web content analysis, allowing users to extract and query information from active webpages to receive context-aware responses. The project distinguishes itself through local AI integration, enabling connections to locally hosted models or private API endpoints to process data without relying on cloud services. It also supports collaborative AI conversations via public sharing links or self-hosted sharing infrastr

This project is an Android RPA framework designed for automating user interfaces and system tasks on rooted Android devices using Python and ADB. It provides a suite of tools for rooted device management, allowing for programmatic control of system settings, application lifecycles, and shell command execution via a remote API. The framework distinguishes itself through a combination of dynamic instrumentation and AI integration. It can inject scripts into running processes to hook Java interfaces and modifies application behavior in real time. Additionally, it supports large language model in

Paddle-Lite is a deep learning inference engine and edge computing runtime designed to execute trained models on mobile and edge devices. It provides a hardware-accelerated inference framework and a decoupled runtime with a minimal binary footprint to operate in resource-constrained environments without third-party dependencies. The project includes a model quantization tool for reducing precision and size via static and dynamic quantization, as well as a computation graph optimizer. These tools reduce latency and memory usage by fusing operators and pruning the model intermediate representat

FluentRead is an open-source browser translation plugin that displays original text alongside its translation directly on web pages. It supports bilingual reading through side-by-side rendering, gesture-triggered activation via mouse hover, double-click, or touch, and keeps all translation data stored locally on the user's device with publicly auditable source code for privacy protection. The plugin offers a multi-engine translation selection, supporting over 20 providers including traditional services and large language models, allowing users to balance accuracy, cost, and privacy for each t

sqlite-vec is a C-based vector library and SQLite extension that adds virtual tables for storing and querying high-dimensional embeddings. It functions as a database plugin for performing nearest neighbor searches using distance metrics such as L2, cosine, and Hamming distance. The project provides a portable embedding store that supports deployment across Android, iOS, desktop environments, and web browsers via WebAssembly. It distinguishes itself by converting numerical arrays into compact binary formats and utilizing quantization to reduce the memory footprint and storage size of vector in

DeepChat is a desktop application that connects to multiple cloud and local AI model providers through a single unified chat interface, while also integrating external ACP-compatible coding and task agents as selectable models. It manages local AI agent sessions with project folders, permission modes, and resumable context for long-running tasks, and connects external tools and data sources via the Model Context Protocol using StreamableHTTP, SSE, or Stdio transports. The application distinguishes itself by supporting remote desktop session control, binding messaging app channels to sessions

Enchanted is a privacy-focused, cross-platform chat frontend for interacting with self-hosted large language models on iOS and macOS. It serves as a native client for communicating with private model servers, specifically providing integration for the Ollama API. The application supports multimodal interactions, allowing users to combine text, image attachments, and voice prompts. It provides tools for local AI model management, including the ability to define persistent system prompts and switch between different models for specific tasks. The interface includes capabilities for rendering m

Enchanted is iOS and macOS app for chatting with private self hosted language models such as Llama2, Mistral or Vicuna using Ollama.

The free AI already on your Mac. CLI tool, OpenAI-compatible server, and interactive chat — all on-device via Apple Intelligence. No API keys, no cloud, no downloads.

AidLearning-Framework is an integrated development platform for building and deploying AI applications on ARM-based devices. It combines Android and Linux operating systems running simultaneously on a single device, providing a unified runtime environment for cross-system AI development. The platform includes hardware acceleration across CPU, GPU, and NPU, with a unified API that automatically selects the optimal compute backend for inference. The framework distinguishes itself by enabling Python-based AI projects to be packaged directly into Android APK files for installation on mobile devic

Serge is a self-hosted web chat interface for running large language models locally using the llama.cpp inference engine. It loads GGUF-format model files directly on your own machine, removing the need for internet connectivity or external API keys, and streams responses to the browser in real time via WebSocket connections. The project is packaged for containerized deployment using Docker and Docker Compose, with a Traefik reverse proxy that handles HTTP and WebSocket routing along with automatic TLS certificate management. Ready-made Kubernetes manifests are also provided, enabling deploym

Dayflow is a privacy-focused productivity tool that records screen activity locally and analyzes it with on-device AI. It captures screen frames at one frame per second, stores everything in a local database, and processes all analysis entirely on the machine to keep data private. The system builds a searchable timeline of work activity and enables natural-language queries about past screen time. The tool distinguishes itself by offering runtime switching between local AI models and cloud providers, allowing users to balance accuracy, privacy, and performance. It automatically runs AI inferen

Electrobun is a desktop application framework and webview-based GUI toolkit used for building cross-platform desktop apps. It provides a TypeScript-based runtime and a native system webview to create interfaces that integrate embedded browser views with host-process logic. The project features a native GPU integration layer with direct FFI bindings, allowing for high-performance GPU surfaces and compute workloads to run within a desktop application. It also includes an inter-process communication bridge using a typed RPC system to exchange data and execute functions between the native backend