✨ ccpocket | Dart
🎯 Primary Use Case:
Enabling developers to remotely control and interact with AI coding agents (like Codex and Claude) from their mobile devices or desktops, facilitating a more flexible and accessible development workflow.
✨ Key Features:
• Control Codex and Claude coding agents from mobile and desktop
• Remote session management and action approval
• Cross-device context persistence
📖 Summary:
CC Pocket is a mobile and desktop application that allows users to control coding agent sessions (Codex and Claude) remotely via a self-hosted WebSocket bridge server. It enables users to start, manage, and approve agent actions from their phone, tablet, or desktop, offering a seamless coding experience across devices.
🔗 Links:
• View Project
• Homepage
================
🔓 Open Source
🎯 Primary Use Case:
Enabling developers to remotely control and interact with AI coding agents (like Codex and Claude) from their mobile devices or desktops, facilitating a more flexible and accessible development workflow.
✨ Key Features:
• Control Codex and Claude coding agents from mobile and desktop
• Remote session management and action approval
• Cross-device context persistence
📖 Summary:
CC Pocket is a mobile and desktop application that allows users to control coding agent sessions (Codex and Claude) remotely via a self-hosted WebSocket bridge server. It enables users to start, manage, and approve agent actions from their phone, tablet, or desktop, offering a seamless coding experience across devices.
🔗 Links:
• View Project
• Homepage
================
🔓 Open Source
🌟 docker-pi-hole | Shell
🎯 Primary Use Case:
To provide a simple and reliable way to deploy Pi-hole as a network-wide ad blocker using Docker.
✨ Key Features:
• Official Pi-hole Docker image
• Network-wide ad blocking
• DNS sinkhole functionality
• Easy deployment with Docker Compose
• Customizable environment variables for configuration
📖 Summary:
This repository provides the official Docker image for Pi-hole, a network-wide ad blocker and DNS sinkhole. It allows users to easily deploy Pi-hole in a containerized environment for efficient ad blocking across their entire network.
🔗 Links:
• View Project
• Homepage
================
🔓 Open Source
🎯 Primary Use Case:
To provide a simple and reliable way to deploy Pi-hole as a network-wide ad blocker using Docker.
✨ Key Features:
• Official Pi-hole Docker image
• Network-wide ad blocking
• DNS sinkhole functionality
• Easy deployment with Docker Compose
• Customizable environment variables for configuration
📖 Summary:
This repository provides the official Docker image for Pi-hole, a network-wide ad blocker and DNS sinkhole. It allows users to easily deploy Pi-hole in a containerized environment for efficient ad blocking across their entire network.
🔗 Links:
• View Project
• Homepage
================
🔓 Open Source
🌟 mgrep | TypeScript
🎯 Primary Use Case:
To provide a natural language interface for searching and retrieving information from local files and the web, particularly useful for developers and AI agents.
✨ Key Features:
• Natural-language search
• Semantic, multilingual, and multimodal search capabilities
• Built-in web search (--web)
• Background indexing with `mgrep watch`
• Friendly device-login flow
📖 Summary:
mgrep is a command-line tool that enables semantic searching across various file types, including code, text, and PDFs, using natural language queries. It features background indexing, web search integration, and is designed for both human and agent use, aiming to reduce token usage for AI agents.
🔗 Links:
• View Project
• Homepage
================
🔓 Open Source
🎯 Primary Use Case:
To provide a natural language interface for searching and retrieving information from local files and the web, particularly useful for developers and AI agents.
✨ Key Features:
• Natural-language search
• Semantic, multilingual, and multimodal search capabilities
• Built-in web search (--web)
• Background indexing with `mgrep watch`
• Friendly device-login flow
📖 Summary:
mgrep is a command-line tool that enables semantic searching across various file types, including code, text, and PDFs, using natural language queries. It features background indexing, web search integration, and is designed for both human and agent use, aiming to reduce token usage for AI agents.
🔗 Links:
• View Project
• Homepage
================
🔓 Open Source
🔥 ESP32-Bit-Pirate | C++
🎯 Primary Use Case:
Hardware hacking and digital protocol analysis
✨ Key Features:
• Interactive CLI via USB Serial or WiFi Web
• Support for numerous protocols (I2C, SPI, UART, 1-Wire, CAN, Wi-Fi, Bluetooth, RFID, etc.)
• Protocol sniffers for I2C, UART, SPI, 1Wire, 2wire, CAN, Wi-Fi, Bluetooth, SubGhz
• EEPROM dump tools
• Scripting using Bus Pirate-style commands
📖 Summary:
ESP32 Bit Pirate is an open-source firmware that transforms an ESP32 device into a versatile hardware hacking tool. It supports a wide array of digital and radio protocols, offering features like sniffing, sending, scripting, and interactive control via a web-based CLI.
🔗 Links:
• View Project
• Homepage
================
🔓 Open Source
🎯 Primary Use Case:
Hardware hacking and digital protocol analysis
✨ Key Features:
• Interactive CLI via USB Serial or WiFi Web
• Support for numerous protocols (I2C, SPI, UART, 1-Wire, CAN, Wi-Fi, Bluetooth, RFID, etc.)
• Protocol sniffers for I2C, UART, SPI, 1Wire, 2wire, CAN, Wi-Fi, Bluetooth, SubGhz
• EEPROM dump tools
• Scripting using Bus Pirate-style commands
📖 Summary:
ESP32 Bit Pirate is an open-source firmware that transforms an ESP32 device into a versatile hardware hacking tool. It supports a wide array of digital and radio protocols, offering features like sniffing, sending, scripting, and interactive control via a web-based CLI.
🔗 Links:
• View Project
• Homepage
================
🔓 Open Source
🚀 app-store-screenshots | TypeScript
🎯 Primary Use Case:
Automating the creation of professional, marketing-ready App Store and Google Play screenshots using AI.
✨ Key Features:
• AI-powered screenshot generation for App Store and Google Play
• Connected canvas editor for seamless element spanning
• Real device frames for iOS, iPad, Android, and Play Store formats
📖 Summary:
This repository provides a Next.js-based editor for generating App Store and Google Play marketing screenshots. It leverages AI coding agents to automate the creation of visually appealing and store-ready screenshot decks, complete with device frames, captions, and connected canvas layouts.
🔗 Links:
• View Project
• Homepage
================
🔓 Open Source
🎯 Primary Use Case:
Automating the creation of professional, marketing-ready App Store and Google Play screenshots using AI.
✨ Key Features:
• AI-powered screenshot generation for App Store and Google Play
• Connected canvas editor for seamless element spanning
• Real device frames for iOS, iPad, Android, and Play Store formats
📖 Summary:
This repository provides a Next.js-based editor for generating App Store and Google Play marketing screenshots. It leverages AI coding agents to automate the creation of visually appealing and store-ready screenshot decks, complete with device frames, captions, and connected canvas layouts.
🔗 Links:
• View Project
• Homepage
================
🔓 Open Source
🚀 agent | Go
🎯 Primary Use Case:
To provide a flexible and scalable software agent for managing and processing video streams from various camera sources for surveillance purposes.
✨ Key Features:
• Open-source video surveillance management agent
• Scalable and isolated component of the Kerberos.io stack
• Supports IP cameras with RTSP H264/H265 streams
• Can process streams from USB cameras, Raspberry Pi cameras, etc. (via camera-to-rtsp)
📖 Summary:
The Kerberos Agent is an open-source, scalable video surveillance management system designed for IP cameras and other video sources. It allows for the processing and management of video streams, enabling features like motion detection and integration into larger surveillance setups.
🔗 Links:
• View Project
• Homepage
================
🔓 Open Source
🎯 Primary Use Case:
To provide a flexible and scalable software agent for managing and processing video streams from various camera sources for surveillance purposes.
✨ Key Features:
• Open-source video surveillance management agent
• Scalable and isolated component of the Kerberos.io stack
• Supports IP cameras with RTSP H264/H265 streams
• Can process streams from USB cameras, Raspberry Pi cameras, etc. (via camera-to-rtsp)
📖 Summary:
The Kerberos Agent is an open-source, scalable video surveillance management system designed for IP cameras and other video sources. It allows for the processing and management of video streams, enabling features like motion detection and integration into larger surveillance setups.
🔗 Links:
• View Project
• Homepage
================
🔓 Open Source
💡 MATRIXON-AGENTIC-BROWSER | HTML
🎯 Primary Use Case:
Smart web automation and AI-assisted browsing using a local, privacy-focused agent.
✨ Key Features:
• Local LLM (GGUF models via llama.cpp)
• AgentPanel sidebar AI assistant
• Web search, page reading, URL navigation tools
• Full-page AI chat with file attachments and LaTeX
• AI-summarized web search results
📖 Summary:
MATRIXON Agentic Browser is an open-source, privacy-focused desktop browser built with Electron. It integrates a fully local AI assistant powered by llama.cpp and GGUF models, enabling smart web automation and interaction without relying on cloud services or API keys.
🔗 Links:
• View Project
================
🔓 Open Source
🎯 Primary Use Case:
Smart web automation and AI-assisted browsing using a local, privacy-focused agent.
✨ Key Features:
• Local LLM (GGUF models via llama.cpp)
• AgentPanel sidebar AI assistant
• Web search, page reading, URL navigation tools
• Full-page AI chat with file attachments and LaTeX
• AI-summarized web search results
📖 Summary:
MATRIXON Agentic Browser is an open-source, privacy-focused desktop browser built with Electron. It integrates a fully local AI assistant powered by llama.cpp and GGUF models, enabling smart web automation and interaction without relying on cloud services or API keys.
🔗 Links:
• View Project
================
🔓 Open Source
❤1🔥1🙏1
🔥 abtop | Rust
🎯 Primary Use Case:
Monitoring and managing the performance and resource utilization of multiple AI coding agent sessions in a terminal environment.
✨ Key Features:
• Real-time monitoring of AI coding agents (Claude Code, Codex CLI, OpenCode)
• Tracks token usage, context window percentage, and rate limits
• Detects open ports and child processes spawned by agents
• Supports multiple active agent sessions across macOS, Linux, and Windows
📖 Summary:
abtop is a terminal-based monitoring tool for AI coding agents, inspired by htop. It provides real-time insights into the performance and resource usage of agents like Claude Code, Codex CLI, and OpenCode, displaying token consumption, rate limits, context window status, and active network ports.
🔗 Links:
• View Project
================
🔓 Open Source
🎯 Primary Use Case:
Monitoring and managing the performance and resource utilization of multiple AI coding agent sessions in a terminal environment.
✨ Key Features:
• Real-time monitoring of AI coding agents (Claude Code, Codex CLI, OpenCode)
• Tracks token usage, context window percentage, and rate limits
• Detects open ports and child processes spawned by agents
• Supports multiple active agent sessions across macOS, Linux, and Windows
📖 Summary:
abtop is a terminal-based monitoring tool for AI coding agents, inspired by htop. It provides real-time insights into the performance and resource usage of agents like Claude Code, Codex CLI, and OpenCode, displaying token consumption, rate limits, context window status, and active network ports.
🔗 Links:
• View Project
================
🔓 Open Source
🚀 docker-android | Shell
🎯 Primary Use Case:
Running Android emulators in a Docker container for automated testing and CI/CD pipelines.
✨ Key Features:
• Minimal Alpine based image with Android emulator and KVM support
• Bundles Java Runtime Environment 11
• Customizable Android version, device type, and image types
• Port-forwarding of emulator and ADB
• Emulator images are wiped on restart
📖 Summary:
This repository provides a minimal and customizable Docker image for running the Android emulator as a service. It's designed for headless execution, making it ideal for CI/CD pipelines and automated testing environments.
🔗 Links:
• View Project
================
🔓 Open Source
🎯 Primary Use Case:
Running Android emulators in a Docker container for automated testing and CI/CD pipelines.
✨ Key Features:
• Minimal Alpine based image with Android emulator and KVM support
• Bundles Java Runtime Environment 11
• Customizable Android version, device type, and image types
• Port-forwarding of emulator and ADB
• Emulator images are wiped on restart
📖 Summary:
This repository provides a minimal and customizable Docker image for running the Android emulator as a service. It's designed for headless execution, making it ideal for CI/CD pipelines and automated testing environments.
🔗 Links:
• View Project
================
🔓 Open Source
🔥 MiroShark | Python
🎯 Primary Use Case:
Simulating and forecasting outcomes of various scenarios using swarm intelligence agents.
✨ Key Features:
• Simulate anything from press releases to policy drafts
• Hundreds of grounded AI agents (Twitter, Reddit, prediction market)
• Hour-by-hour simulation progression
• Interactive simulation: chat with agents, drop breaking news, fork timelines
• Generates reports with cited posts and trades
📖 Summary:
MiroShark is a universal swarm intelligence engine that simulates complex scenarios by spawning hundreds of AI agents. These agents interact within a simulated environment, posting, arguing, and trading based on an initial input, providing insights and reports on the simulated outcomes.
🔗 Links:
• View Project
• Homepage
================
🔓 Open Source
🎯 Primary Use Case:
Simulating and forecasting outcomes of various scenarios using swarm intelligence agents.
✨ Key Features:
• Simulate anything from press releases to policy drafts
• Hundreds of grounded AI agents (Twitter, Reddit, prediction market)
• Hour-by-hour simulation progression
• Interactive simulation: chat with agents, drop breaking news, fork timelines
• Generates reports with cited posts and trades
📖 Summary:
MiroShark is a universal swarm intelligence engine that simulates complex scenarios by spawning hundreds of AI agents. These agents interact within a simulated environment, posting, arguing, and trading based on an initial input, providing insights and reports on the simulated outcomes.
🔗 Links:
• View Project
• Homepage
================
🔓 Open Source
💡 devFind | JavaScript
🎯 Primary Use Case:
To serve as a discoverable directory for developers to showcase their skills and find collaborators or employment opportunities.
✨ Key Features:
• Search functionality for developers based on skills, location, or name
• Platform for developers to showcase skills and connect with collaborators
• User-friendly and searchable format for discovering developers
• Developer profiles created in JSON format
📖 Summary:
devFind is an open-source platform designed for developers to create profiles showcasing their skills and connect with potential collaborators. Profiles are created in JSON format and are discoverable through a user-friendly, searchable web interface.
🔗 Links:
• View Project
• Homepage
================
🔓 Open Source
🎯 Primary Use Case:
To serve as a discoverable directory for developers to showcase their skills and find collaborators or employment opportunities.
✨ Key Features:
• Search functionality for developers based on skills, location, or name
• Platform for developers to showcase skills and connect with collaborators
• User-friendly and searchable format for discovering developers
• Developer profiles created in JSON format
📖 Summary:
devFind is an open-source platform designed for developers to create profiles showcasing their skills and connect with potential collaborators. Profiles are created in JSON format and are discoverable through a user-friendly, searchable web interface.
🔗 Links:
• View Project
• Homepage
================
🔓 Open Source
🚀 awesome-osint-arsenal | Shell
🎯 Primary Use Case:
To provide a centralized, easy-to-install, and organized collection of cybersecurity and OSINT tools for security professionals, researchers, and ethical hackers.
✨ Key Features:
• 751+ OSINT & Security Tools
• 50+ Categories
• One-command installer (install.sh)
• Categorized installation scripts (osint.sh, redteam.sh, etc.)
📖 Summary:
Awesome OSINT Arsenal is a comprehensive shell script toolkit designed for Kali Linux and other Debian-based systems, offering over 750 tools categorized for various cybersecurity tasks. It simplifies the installation and access to tools for reconnaissance, red teaming, blue teaming, digital forensics, hardware analysis, and practice labs, with a special subset for Termux on Android.
🔗 Links:
• View Project
================
🔓 Open Source
🎯 Primary Use Case:
To provide a centralized, easy-to-install, and organized collection of cybersecurity and OSINT tools for security professionals, researchers, and ethical hackers.
✨ Key Features:
• 751+ OSINT & Security Tools
• 50+ Categories
• One-command installer (install.sh)
• Categorized installation scripts (osint.sh, redteam.sh, etc.)
📖 Summary:
Awesome OSINT Arsenal is a comprehensive shell script toolkit designed for Kali Linux and other Debian-based systems, offering over 750 tools categorized for various cybersecurity tasks. It simplifies the installation and access to tools for reconnaissance, red teaming, blue teaming, digital forensics, hardware analysis, and practice labs, with a special subset for Termux on Android.
🔗 Links:
• View Project
================
🔓 Open Source
🌟 kadr | TypeScript
🎯 Primary Use Case:
AI-assisted video editing with real-time GPU acceleration and integrated AI coding for motion graphics.
✨ Key Features:
• Full multi-track editing (video/audio/text)
• GPU compositing (WebGL2) for real-time preview and rendering
• Keyframing for various properties (position, scale, rotation, opacity, masks)
• True 3D capabilities for clips and tracks
• Animated masks with multiple shapes and feathering
📖 Summary:
Kadr is an AI-native, GPU-accelerated video editor built with Electron, React, and TypeScript. It integrates AI agents directly into the timeline, allowing for real-time editing assistance and automated features like subtitle generation and motion graphics creation.
🔗 Links:
• View Project
================
🔓 Open Source
🎯 Primary Use Case:
AI-assisted video editing with real-time GPU acceleration and integrated AI coding for motion graphics.
✨ Key Features:
• Full multi-track editing (video/audio/text)
• GPU compositing (WebGL2) for real-time preview and rendering
• Keyframing for various properties (position, scale, rotation, opacity, masks)
• True 3D capabilities for clips and tracks
• Animated masks with multiple shapes and feathering
📖 Summary:
Kadr is an AI-native, GPU-accelerated video editor built with Electron, React, and TypeScript. It integrates AI agents directly into the timeline, allowing for real-time editing assistance and automated features like subtitle generation and motion graphics creation.
🔗 Links:
• View Project
================
🔓 Open Source
💡 homer | Vue
🎯 Primary Use Case:
A self-hosted, customizable dashboard for organizing and quickly accessing frequently used web services and applications.
✨ Key Features:
• Lightweight & Fast
• Low / No maintenance
• Simple YAML file configuration
• Installable (PWA)
• Smart cards
📖 Summary:
Homer is a lightweight, self-hosted static homepage designed to provide a simple and fast dashboard for managing your server's services. It is configured via a YAML file and offers features like smart cards, fuzzy search, and theme customization.
🔗 Links:
• View Project
• Homepage
================
🔓 Open Source
🎯 Primary Use Case:
A self-hosted, customizable dashboard for organizing and quickly accessing frequently used web services and applications.
✨ Key Features:
• Lightweight & Fast
• Low / No maintenance
• Simple YAML file configuration
• Installable (PWA)
• Smart cards
📖 Summary:
Homer is a lightweight, self-hosted static homepage designed to provide a simple and fast dashboard for managing your server's services. It is configured via a YAML file and offers features like smart cards, fuzzy search, and theme customization.
🔗 Links:
• View Project
• Homepage
================
🔓 Open Source
🔥 penpot | Clojure
🎯 Primary Use Case:
Collaborative design and development of digital products, bridging the gap between designers and developers with open standards and self-hosting capabilities.
✨ Key Features:
• Open-source design and prototyping platform
• Real-time collaboration
• Self-hosting for full infrastructure ownership
• Works with open standards (SVG, CSS, HTML, JSON)
📖 Summary:
Penpot is an open-source design and prototyping platform that enables teams to collaborate on digital product development. It emphasizes full ownership of design infrastructure through self-hosting and works with open standards like SVG, CSS, and HTML, integrating design and code with features like Design Tokens and a powerful API.
🔗 Links:
• View Project
• Homepage
================
🔓 Open Source
🎯 Primary Use Case:
Collaborative design and development of digital products, bridging the gap between designers and developers with open standards and self-hosting capabilities.
✨ Key Features:
• Open-source design and prototyping platform
• Real-time collaboration
• Self-hosting for full infrastructure ownership
• Works with open standards (SVG, CSS, HTML, JSON)
📖 Summary:
Penpot is an open-source design and prototyping platform that enables teams to collaborate on digital product development. It emphasizes full ownership of design infrastructure through self-hosting and works with open standards like SVG, CSS, and HTML, integrating design and code with features like Design Tokens and a powerful API.
🔗 Links:
• View Project
• Homepage
================
🔓 Open Source
🌟 web-to-app | Kotlin
🎯 Primary Use Case:
To allow users to easily convert any website or web project into a fully functional, installable Android application directly from their mobile device.
✨ Key Features:
• On-device APK building for websites and local runtimes
• Supports various input types: URL, project folder, media library
• Customizable WebView, toolbar, splash screen, permissions, and signing
• Extensible with JS/CSS modules and userscripts
📖 Summary:
WebToApp is an on-device Android application that transforms websites, HTML projects, and various local runtimes into standalone Android APKs. It offers a comprehensive suite of tools for building, customizing, and managing these apps directly on your phone without relying on remote services.
🔗 Links:
• View Project
================
🔓 Open Source
🎯 Primary Use Case:
To allow users to easily convert any website or web project into a fully functional, installable Android application directly from their mobile device.
✨ Key Features:
• On-device APK building for websites and local runtimes
• Supports various input types: URL, project folder, media library
• Customizable WebView, toolbar, splash screen, permissions, and signing
• Extensible with JS/CSS modules and userscripts
📖 Summary:
WebToApp is an on-device Android application that transforms websites, HTML projects, and various local runtimes into standalone Android APKs. It offers a comprehensive suite of tools for building, customizing, and managing these apps directly on your phone without relying on remote services.
🔗 Links:
• View Project
================
🔓 Open Source
✨ mcp-ssh-manager | JavaScript
🎯 Primary Use Case:
Enabling AI models to automate and manage remote server infrastructure through a unified protocol.
✨ Key Features:
• AI-powered remote server management via Model Context Protocol (MCP)
• Support for Claude Code and OpenAI Codex
• Execute commands on remote SSH servers
• File transfer capabilities
• Database management operations
📖 Summary:
MCP SSH Manager is a Node.js server that allows AI models like Claude Code and OpenAI Codex to manage remote SSH servers. It provides a framework for AI to execute commands, transfer files, manage databases, perform backups, and monitor server health, enabling sophisticated DevOps automation.
🔗 Links:
• View Project
• Homepage
================
🔓 Open Source
🎯 Primary Use Case:
Enabling AI models to automate and manage remote server infrastructure through a unified protocol.
✨ Key Features:
• AI-powered remote server management via Model Context Protocol (MCP)
• Support for Claude Code and OpenAI Codex
• Execute commands on remote SSH servers
• File transfer capabilities
• Database management operations
📖 Summary:
MCP SSH Manager is a Node.js server that allows AI models like Claude Code and OpenAI Codex to manage remote SSH servers. It provides a framework for AI to execute commands, transfer files, manage databases, perform backups, and monitor server health, enabling sophisticated DevOps automation.
🔗 Links:
• View Project
• Homepage
================
🔓 Open Source
✨ FreeDomain | HTML
🎯 Primary Use Case:
To offer free domain names, enabling individuals and organizations to establish an online presence without financial barriers.
✨ Key Features:
• Free domain name registration
• Multiple available domain extensions (.DPDNS.ORG, .US.KG, .QZZ.IO, .XX.KG, .QD.JE)
• Compatibility with popular DNS providers (Cloudflare, Afraid.org, Hostry)
• User-friendly dashboard for domain management
• Community support via Discord
📖 Summary:
DigitalPlat FreeDomain provides free domain names to individuals and organizations, aiming to make digital identity accessible to everyone. Users can register unique domain extensions and connect them to their preferred DNS providers.
🔗 Links:
• View Project
• Homepage
================
🔓 Open Source
🎯 Primary Use Case:
To offer free domain names, enabling individuals and organizations to establish an online presence without financial barriers.
✨ Key Features:
• Free domain name registration
• Multiple available domain extensions (.DPDNS.ORG, .US.KG, .QZZ.IO, .XX.KG, .QD.JE)
• Compatibility with popular DNS providers (Cloudflare, Afraid.org, Hostry)
• User-friendly dashboard for domain management
• Community support via Discord
📖 Summary:
DigitalPlat FreeDomain provides free domain names to individuals and organizations, aiming to make digital identity accessible to everyone. Users can register unique domain extensions and connect them to their preferred DNS providers.
🔗 Links:
• View Project
• Homepage
================
🔓 Open Source
🚀 FastVideo | Python
🎯 Primary Use Case:
Accelerated video generation and real-time inference for diffusion models.
✨ Key Features:
• End-to-end post-training support for bidirectional and autoregressive models
• Data preprocessing pipeline for video, image, and text data
• Distribution Matching Distillation (DMD2) stepwise distillation
• Sparse attention with Video Sparse Attention
• Sparse distillation for >50x denoising speedup
📖 Summary:
FastVideo is a unified framework for accelerating video generation through post-training techniques and optimized inference. It supports various training methods like finetuning and distillation, and offers significant performance improvements for real-time video generation and editing.
🔗 Links:
• View Project
• Homepage
================
🔓 Open Source
🎯 Primary Use Case:
Accelerated video generation and real-time inference for diffusion models.
✨ Key Features:
• End-to-end post-training support for bidirectional and autoregressive models
• Data preprocessing pipeline for video, image, and text data
• Distribution Matching Distillation (DMD2) stepwise distillation
• Sparse attention with Video Sparse Attention
• Sparse distillation for >50x denoising speedup
📖 Summary:
FastVideo is a unified framework for accelerating video generation through post-training techniques and optimized inference. It supports various training methods like finetuning and distillation, and offers significant performance improvements for real-time video generation and editing.
🔗 Links:
• View Project
• Homepage
================
🔓 Open Source
❤1
🚀 koharu | Rust
🎯 Primary Use Case:
Automated, local-first manga translation with ML-powered features and editable export options.
✨ Key Features:
• Automatic detection of text regions, speech bubbles, and cleanup masks
• OCR for manga dialogue, captions, and other page text
• Inpainting to remove source lettering from the page
• Translation with local or remote LLM backends
📖 Summary:
Koharu is a Rust-based desktop application that automates the process of translating manga. It leverages machine learning models for text detection, OCR, inpainting to remove original text, and translation via LLMs, all running locally for privacy. The application supports advanced text rendering and offers features like layered PSD export for further editing.
🔗 Links:
• View Project
• Homepage
================
🔓 Open Source
🎯 Primary Use Case:
Automated, local-first manga translation with ML-powered features and editable export options.
✨ Key Features:
• Automatic detection of text regions, speech bubbles, and cleanup masks
• OCR for manga dialogue, captions, and other page text
• Inpainting to remove source lettering from the page
• Translation with local or remote LLM backends
📖 Summary:
Koharu is a Rust-based desktop application that automates the process of translating manga. It leverages machine learning models for text detection, OCR, inpainting to remove original text, and translation via LLMs, all running locally for privacy. The application supports advanced text rendering and offers features like layered PSD export for further editing.
🔗 Links:
• View Project
• Homepage
================
🔓 Open Source