📦 soul-ailab/soulx-flashtalk

SoulX-FlashTalk: Real-Time Infinite Streaming of Audio-Driven Avatars

Generate endless, real-time video of a person talking directly from audio input. SoulX-FlashTalk solves the slow startup problem for large video models by launching in under a second. It streams video frames at thirty-two frames per second on high-end hardware, creating avatars that appear to talk forever without breaking. The key feature is its ability to correct itself as it generates, ensuring the face stays stable and lifelike throughout long conversations. This makes it perfect for live digital humans that need to feel immediate and present. You get a fast, continuous talking head without the usual lag, making it a standout tool for real-time interaction.

🆔 @hackernewsgithubprojects

📦 lyehe/build_gpu_colmap

COLMAP GPU Builds: No Install Hell

Building COLMAP for GPU acceleration usually turns into a nightmare of compiling dependencies and fighting drivers, but this project hands you ready-made binaries that just work. It packages the photo-to-3D tool with NVIDIA support so you can skip the build scripts entirely. On Windows, the files are huge because they include everything needed to run, meaning you don't have to install the CUDA toolkit separately. On Linux, they stay small by relying on your system libraries. You get fast reconstruction speeds without the usual setup headaches. It is the shortcut for anyone who wants results, not a computer science degree. Save yourself the pain and grab a pre-built package today.

🆔 @hackernewsgithubprojects

📦 nb-corp/gpt-with-taste

GPT with Taste: Better Judgment, Less Noise

GPT with Taste makes your AI coding assistant act with actual judgment instead of just guessing. It drops a special recipe into OpenAI Codex to stop the model from burying your results in defensive disclaimers and unnecessary process talk. The standout feature is how it handles getting-started guides. Instead of leading with environment checks and failure handling, it hands you a working example and tells you exactly what to click to see it change. It cuts the ceremony and gets you to the actual code faster. It is a simple fix for AI that talks too much and builds too little. Try it out and see the difference.

🆔 @hackernewsgithubprojects

📦 chenglou/handwired

Handwired: JavaScript as a Neural Net

Handwired turns regular JavaScript code into a neural net without any training. It converts standard logic like loops and branches into simple math that runs exactly like your original code. This lets you search a hundred thousand documents in milliseconds by using your graphics processor to parallelize tasks normally done one at a time. The result is a fully differentiable program that acts like a neural network but needs no data to learn from. This approach makes it possible to attach specific logic to larger machine learning models for end-to-end training. It’s a clever way to merge traditional programming with deep learning for high-speed, flexible applications.

🆔 @hackernewsgithubprojects

📦 crocswap/integer-mult-bounds

integer-mult-bounds

A crowd of independent researchers just managed to shave a tiny sliver off the theoretical speed limit of multiplying giant numbers. This project, called integer mult bounds, isn't about making your phone faster right now; it is about proving that big number math can be done with just a hair fewer steps than we thought possible. It solves a deep puzzle in computer science by building a massive, complex network of calculations that reuses parts of the work to save energy. This is pure math, not a quick fix, but it shows how a global community can collaborate to push the boundaries of what is computationally possible.

🆔 @hackernewsgithubprojects

📦 lucky-verma/mwb-linux

Control Windows from Linux with Mouse Without Borders

Move your mouse to the edge of your Linux screen, and the cursor instantly hops over to your Windows PC, taking your keyboard and clipboard along for the ride. This project, called Mouse Without Borders for Linux, solves the annoying problem of needing two separate sets of input devices when you run both operating systems. It acts like a magical bridge, letting you copy text on one machine and paste it on the other without touching a USB cable. The coolest part is that your Linux mouse stays perfectly still on the desktop while your digital avatar moves on Windows, preventing any accidental clicks.

📰 https://news.ycombinator.com/item?id=48823108

🆔 @hackernewsgithubprojects

📦 nutlope/gloss

Gloss: The Reader That Answers In The Margin

Gloss is the reading tool that finally lets you talk back to what you are reading. Instead of switching to a search bar, you simply highlight a sentence and type your question. The artificial intelligence answers right there in the margin, pointing to the specific paragraphs it used. It works on articles, books, and PDFs. The design feels like a physical notebook, with the text on the left and the AI notes on the right. It keeps your notes and highlights safe on your device, so you do not need an account to start.

🆔 @hackernewsgithubprojects

📦 drowzeys/keys-wasserman-slate-ai-studio-powered-by-tensorfold-mac-dgxspark

Keys Wasserman Slate AI Studio: Your Personal Film Crew

A single Mac can now function as a complete film production crew. Keys Wasserman Slate AI Studio merges a professional planning workflow with high-speed AI rendering, letting you script scenes, cast characters, and generate finished video clips with sound directly on your machine. The platform handles the entire pipeline by converting detailed text descriptions into visual assets using specialized models, effectively automating the transition from concept to final cut without needing a massive server cluster. It is a practical demonstration of how local hardware can manage complex creative tasks that once required cloud infrastructure, giving creators total control over their digital assets.

🆔 @hackernewsgithubprojects

📦 init-co-jp/diff-tour

How to Read Code Like a Senior Engineer

A command line tool now turns your git changes into a guided tour that mimics how a senior engineer actually reviews code. Instead of just staring at a wall of text, you step through a narrative that explains the purpose, points out risky spots, and asks you to predict outcomes before revealing the answer. It uses an AI model you already have to read your code, but it locks the AI down so it can only look and never touch your files. This is perfect for understanding complex refactors or reviewing AI-generated code where the logic is not obvious.

🆔 @hackernewsgithubprojects

📦 mertkayacs/reevesagents

ReevesAgents: The AI Team Manager

ReevesAgents is the local workspace tool that finally lets your favorite AI coding assistants actually work together on the same project. Instead of juggling separate windows or manually copy-pasting code between different tools, it spins up a shared terminal environment where you can launch multiple agents side-by-side instantly. The really wild part is that you can let one agent act as the boss, giving it control to spawn, steer, and stop the other workers directly. It’s like hiring a small team of AI developers and handing them a shared office, all running safely on your own machine without any cloud dependency.

🆔 @hackernewsgithubprojects

📦 crosstalk-solutions/unifi-toolkit

House Arrest: One-Click Network Lockdowns for UniFi

UniFi owners can now lock down any device or entire network with a single reversible click, a capability that transforms how home networks handle security. This project, called UI Toolkit, adds a new tool named House Arrest that uses your router’s built-in firewall to instantly cut off internet or local access for specific devices. It is designed to be completely honest about exactly what is being blocked, showing you a clear preview before you apply any changes and ensuring that releasing the lockdown removes only the rules it created. This means you can safely isolate a rogue phone or a guest network without permanently breaking your setup.

🆔 @hackernewsgithubprojects

📦 dimthink/priceai

PriceAI: Compare 200+ AI Subscription Deals

PriceAI hunts down the cheapest way to pay for your favorite AI tools. It scans over two hundred different sellers to find the lowest price for subscriptions like ChatGPT Plus or Claude Pro. You can instantly compare costs, check if stock is available, and jump straight to the official store. This tool cuts through the confusion of regional pricing to show you exactly where to buy. Stop overpaying and get the best deal on your next subscription.

🆔 @hackernewsgithubprojects

📦 segfault42/heliosgen

HeliosGen: Your Local AI Art Studio

Build visual pipelines for AI images and videos right on your own desktop. HeliosGen is a free, self-hosted app that replaces expensive subscription services with a simple local tool. You drag, drop, and connect nodes on an infinite canvas to chain together different AI models, reference photos, and automations. The coolest part is that everything stays on your machine, so your data never leaves your computer and you avoid monthly fees or expiring credits. You just plug in your own API key and start creating. It is the perfect way to experiment with complex generative workflows without vendor lock-in or cloud dependencies.

🆔 @hackernewsgithubprojects

📦 ertasselli/openterminal

OpenTerminal: The Free Trading Dashboard

Build a professional trading workspace on your own computer without spending two thousand dollars a month. OpenTerminal gives you the dark, dense, keyboard-driven interface traders pay for, but it runs entirely on free public market data. You get live charts, stock quotes, and a sector heatmap all in one place. The real magic is that it stitches together several free data sources so if one fails, another takes over instantly. There are no API keys to manage and no subscriptions to pay. Just clone the code and run it to get a fully functional terminal in under a minute.

🆔 @hackernewsgithubprojects

📦 stoopwenpuenq/jackie-fmri-sim

Jackie FMRI Sim: The Dog Brain Twin

Jackie FMRI Sim is the research prototype that turns a single dog’s brain scans into a live emotional model. Built from real functional MRI, EEG, and heart rate data collected over six different social situations, the project maps how a dog named Jackie reacts to everything from a new street dog to his best friend. You can type in a stimulus, like the smell of chicken or a specific social encounter, and the system predicts a six-part emotional vector alongside heart rate changes and brain activity.

🆔 @hackernewsgithubprojects

📦 c9dev/penguin-mail

Penguin Mail: The Local-First Email Client

Take control of your email with Penguin Mail, a desktop client for Linux and macOS that keeps your data private. It connects directly to Gmail, Outlook, or IMAP accounts, so your mail stays on your computer rather than syncing through a third-party server. The standout feature here is the built-in AI assistant. It runs locally on your own machine, letting you summarize long threads, draft replies, or find free time in your calendar without your data ever leaving your device. You also get full encryption support and calendar integration in one sleek app. If you want a fast, private, and intelligent email experience, this is the tool to try.

🆔 @hackernewsgithubprojects

📦 vyrx-dev/toofan

toofan: Type real code in your terminal

Toofan is the terminal typing test that lets you practice with actual code snippets. Most typing sites force you to copy random English words, but this tool hands you real, hand-written exercises in languages like Python or Go. It runs entirely offline inside your terminal, so there is no account, no browser, and no data leaving your machine. You get live speed tracking and a personal dashboard to watch your progress grow. If you code for a living, this is the fastest way to keep your fingers sharp without opening a new tab.

🆔 @hackernewsgithubprojects

📦 vallierdev/uu-remote-linux

Control Your Linux Box From Anywhere

Turn any Linux machine into a remote desktop you can control from your phone, Mac, or Windows PC without buying new hardware. NetEase UU Remote is a great tool, but they skipped Linux entirely. This project fixes that by writing a native host that plugs right into the official clients. The coolest part? It makes your Linux box appear in your device list as if it were a Windows PC, letting you stream your screen, move the mouse, and even grab the terminal. It handles audio, file transfers, and port mapping over a secure connection. No Wine, no hacks, just a clean Rust build that works.

🆔 @hackernewsgithubprojects

📦 vstorm-co/agenticos

AgenticOS: Your Company's AI, Your Rules

Run your entire team of AI workers on your own servers, not someone else's. AgenticOS gives you a private control center where you define exactly what those agents can do, how much they can spend, and who has to approve their actions. You get a full audit trail of every decision, so nothing happens in the dark. It connects to thousands of existing tools, letting you build specific workers for legal, healthcare, or logistics without leaving your own infrastructure. If you need AI that follows your company's strict rules, this is how you keep it safe.

📰 https://news.ycombinator.com/item?id=49705577

🆔 @hackernewsgithubprojects

📦 supermedo/uncharted_pc

Uncharted PC: PS4 Games on Windows

Play the complete Nathan Drake collection of Uncharted directly on your Windows computer without needing a PlayStation console. This project runs the original PlayStation four game code natively on your PC by translating graphics into Vulkan, ensuring you get the exact same experience as the console. It handles the tricky job of making old hardware instructions work on modern graphics cards, so you can enjoy all three adventures with modern upscaling tech like DLSS and high refresh rates. No emulation layer or heavy bloat, just the raw game files running on your machine.

🆔 @hackernewsgithubprojects

📦 tykimos/agent-native-agent

Agent-Native Agent: The App That Rewrites Itself While You Watch

Agent-Native Agent is a self-hosted dashboard where the coding agent is literally the engine running your application. Instead of just chatting with a bot, you watch the live interface and talk to the agent, which then proposes and applies code changes on the spot. This means the app evolves in real time as you use it, eliminating the usual delay between asking for a feature and seeing it work. It’s built specifically for Claude Code and Model Context Protocol, letting you operate a living system that changes its own rules based on your conversation.

🆔 @hackernewsgithubprojects