📦 xiaomi-research/spatio-lm

SpatioLM: Spatial Intelligence for Vision-Language Models

SpatioLM lets AI models truly understand physical space in images and videos. It takes existing vision-language models and adds a lightweight module that teaches them about depth and 3D structure without needing special hardware. By learning from a teacher model that understands geometry, SpatioLM helps the AI answer questions like which object is closer or how things are arranged in a room. It works on standard photos and video clips, offering both reasoning skills and precise depth perception. This makes complex spatial tasks much easier for your favorite AI assistants to handle.

🆔 @hackernewsgithubprojects

📦 syhya/mlsys26-flashinfer-contest

How AI Agents Optimized CUDA Kernels for MLSys 2026

This is the MLSys 2026 FlashInfer Contest package that shows how AI coding agents can autonomously optimize complex computer chip instructions. It documents two winning approaches for generating high-speed graphics processing unit kernels. One method involves a human guiding the AI with specific skills, while the other lets an agent system work completely alone, planning and executing code without any human interference. The results are impressive, landing in the top three across multiple competition categories. You can see the exact step-by-step logs of how the AI solved these challenging coding tasks, offering a fascinating glimpse into the future of automated software engineering.

🆔 @hackernewsgithubprojects

📦 klboke/kkrepo

kkRepo: Your Self-Hosted Nexus Replacement

kkRepo lets you ditch the annoying limits of Sonatype Nexus by hosting every type of software package in one place. It supports Maven, npm, Python, Docker, Rust, Go, Helm, and many more, acting as a direct, compatible replacement for your existing setup. The real magic is the one-click migration tool that moves your users, permissions, and packages over without breaking a sweat, so your CI pipelines keep working exactly as they did before. It runs fast with low memory and handles large-scale storage via S3 or local files. If you are tired of component caps or slow upgrades, this is the clean, open-source fix you have been looking for.

📰 https://news.ycombinator.com/item?id=49179927

🆔 @hackernewsgithubprojects

📦 zyc-hust/csnet

CSNet Multi-Focus Image Fusion

CSNet is the machine learning tool that finally makes blurry photos sharp without guessing. It tackles multi-focus image fusion by intelligently picking the clearest parts of several shots and blending them into one crisp picture. Think of it as a digital camera that knows exactly which details are in focus, even when the lighting is tricky. The project lets developers run this process using standard Python and PyTorch, calculating key quality scores to prove the images look better. While some heavy calculations still rely on MATLAB, the core engine is ready to go.

🆔 @hackernewsgithubprojects

📦 halcyonnouveau/soppo

Soppo: Go With Better Safety

Write Go code that catches nil errors and missing enum cases before they crash your app. Soppo is a language that translates your code into standard Go, but adds powerful safety features Go lacks. You get automatic nil safety checks that stop pointer crashes at compile time, and pattern matching that ensures you handle every possible outcome. It also brings clean error handling and immutable data structures to keep your logic predictable. Since it compiles directly to Go, you can still use any existing Go library without changing your workflow. It’s basically Go with the safety net you wish it had from the start.

📰 https://news.ycombinator.com/item?id=49180168

🆔 @hackernewsgithubprojects

📦 nimi-research/tycho

Tycho: AI That Learns by Building World Models

Solve complex puzzles by building its own mental models of the game world. Tycho is an AI agent for the ARC AGI challenge that refuses to guess. Instead of just looking at pixels, it watches how things move, then writes a small Python script to describe exactly how the universe works. It tests this new understanding against what it sees, checking if its own invented rules hold true. When the model works, Tycho uses it to plan moves and solve problems it has never seen before.

🆔 @hackernewsgithubprojects

📦 jgraph/drawio-mcp

drawio-mcp

Artificial intelligence models used to struggle with visual design, but this project changes that by letting language models actually build diagrams. Draw.io MCP Server connects your favorite AI assistants directly to the popular draw.io editor. Instead of just describing a flowchart or system architecture in text, the AI generates the actual diagram files for you. You can create interactive visuals right inside your chat or open them in the full editor for editing. It supports complex shapes from cloud providers and standard diagrams alike. This makes explaining technical concepts or documenting software much easier. Your diagrams stay local, keeping your data secure.

📰 https://news.ycombinator.com/item?id=49180521

🆔 @hackernewsgithubprojects

📦 sandai-org/magi-2-preview

Magi-2 Preview: Video With Sound

Create ten-second videos complete with synchronized sound just by typing a description or uploading a single still image. Magi-2 Preview acts like a digital film studio that uses a massive neural network to dream up scenes while simultaneously composing the audio track you hear. It handles everything from text to video or image to video, stitching the picture and the noise into one polished clip. The real magic is that it generates the audio alongside the visuals, so you get a full sensory experience without needing separate tools. It is a clever step toward making content creation faster and more intuitive for everyone.

🆔 @hackernewsgithubprojects

📦 copilotkit/channels-sdk

Channels SDK

CopilotKit Channels SDK turns any chat app into a conversation with buttons and charts. You write the logic once and it works in Slack, Teams, Discord, or Telegram. The agent shows native buttons, lets you click, and pauses for your approval before acting. Your code stays yours while they handle the platform connection. It is a simple way to bring AI into your daily chat without leaving your usual apps.

🆔 @hackernewsgithubprojects

📦 ryanzhou/deepseek-v4-flash-mi300x

DeepSeek V4 Flash MI300X: The Patch That Runs In Production

DeepSeek V4 Flash MI300X is not a new software build but a precise collection of production patches that fix specific performance bottlenecks in the vLLM inference engine for AMD graphics cards. The repository contains only a readme and a folder of python files that act as direct replacements for internal components. These files are byte-for-byte identical to what actually runs in real servers today. The project solves the problem of getting large language models to run efficiently on specialized hardware without waiting for upstream updates. Instead of rewriting everything, the developer surgically tweaks specific mathematical operations and memory handling routines to squeeze out more speed.

📰 https://news.ycombinator.com/item?id=49166386

🆔 @hackernewsgithubprojects

📦 faramesh/faramesh-core

Stop AI Agents From Messing Up: Faramesh-Core

Faramesh-core lets you write simple rules that stop AI agents from making risky moves. Instead of trusting the model to behave, a local daemon checks every single action first. You define who can do what, like blocking account cancellations or limiting email sends. It enforces these limits strictly without needing cloud services. You get a tamper-proof log of every decision, so you know exactly what the agent did and why. It works with most agent tools you already use. This gives you real control over automated workflows, ensuring safety and clear accountability from day one.

📰 https://news.ycombinator.com/item?id=48023859

🆔 @hackernewsgithubprojects

📦 weavemindai/weft

Weft: The AI Programming Language

Weft is the programming language that finally lets AI systems pause for human input without writing a single line of webhook code. Think of it as a way to build complex software where artificial intelligence, databases, and actual people are all just equal ingredients. Instead of struggling to manage state or poll for answers, you simply drop a human query node into your workflow. The program stops, sends a request to a user, and then wakes up exactly where it left off days later, ready to continue. It handles the waiting automatically, so you can focus on the logic rather than the plumbing.

🆔 @hackernewsgithubprojects

📦 melgarafael/deskcommcrm

DeskcommCRM: The AI-Powered Chat Sales Tool

DeskcommCRM lets you build and run your own sales robots that actually talk to customers on WhatsApp without you writing any code. It is a self-hosted system that turns your phone into a full customer service hub, handling leads, tracking conversations, and managing your sales pipeline all in one place. The real magic is in the built-in AI agents that can chat with buyers, answer questions from your knowledge base, and even draft proposals for you. You keep total control over your data while getting the power of automation that usually costs a fortune.

🆔 @hackernewsgithubprojects

📦 kageroumado/rilmazafone

rilmazafone

rilmazafone is the visual designer that lets you paint your Mac installer instead of writing code. You know how annoying it is to manually arrange icons on a desktop so they look tidy when someone downloads your app? This tool fixes that. You simply drag your app icon and any other files onto a canvas, snap them exactly where you want them, and style the background with gradients or photos. It handles all the messy binary magic in the background to generate a perfectly laid out disk image file with a single click. No terminal commands, no guesswork. Just design it, build it, and ship it.

📰 https://news.ycombinator.com/item?id=48140887

🆔 @hackernewsgithubprojects

📦 ben-z/findphone

Find Phone Without Find My

It turns your Mac into a handheld signal detector for nearby Bluetooth gadgets when Apple’s Find My service is blocked. This project scans for devices by signal strength and lets you hunt for a specific one by name. You get a visual bar and optional audio clicks that speed up as you walk closer, letting you sweep a room by ear instead of staring at a screen. It works by reading Bluetooth data directly from the command line, which is handy when a device is managed by school or work policies that disable standard tracking tools.

📰 https://news.ycombinator.com/item?id=49181584

🆔 @hackernewsgithubprojects

📦 helloianneo/ian-xiaohei-illustrations

Turn Chinese Articles into Hand-drawn Illustrations with Ian Xiaohei

Ian Xiaohei Illustrations lets you turn dense Chinese articles into clean, hand-drawn visuals by using an AI agent that spots a single key idea—like a workflow, a metaphor, or a decision point—and builds a 16 by 9 illustration around it. Instead of generic clip art, it features a recurring character named Xiaohei, a deadpan black figure with dot eyes, doing the actual conceptual work. The style stays minimalist: pure white backgrounds, sketchy black lines, and sparse red, orange, or blue Chinese notes. You paste your text, and the tool outputs a shot list of images where Xiaohei explains the concept, not just decorates the margin.

📰 https://news.ycombinator.com/item?id=49182451

🆔 @hackernewsgithubprojects

📦 firecrawl/anydoc

Anydoc: Convert any office document to Markdown

anydoc turns Word, PowerPoint, Excel, PDF, and more into clean Markdown. Built in Rust, it converts files in single-digit milliseconds while keeping tables, lists, and formatting intact. It includes bindings for Node.js, Python, and the browser, so you can integrate document conversion directly into your apps. Unlike slower tools that struggle with mixed formats, anydoc handles everything from old.doc files to modern.xlsx spreadsheets with one consistent output. It even works locally in your browser via WebAssembly, meaning your documents never leave your machine. For developers building AI agents, it ships as a ready-to-use skill, making it easy to teach tools how to read office files.

🆔 @hackernewsgithubprojects

📦 davidwhitney/jsxcore

JsxCore: React Views Inside.NET

JsxCore is a view engine that lets you write React components directly inside ASP.NET Core. It solves the headache of mixing C# back ends with separate JavaScript front ends by letting you return JSX files straight from your controllers. You get hot reload and automatic type checking without installing Node or npm, because the library handles the heavy lifting itself. It even generates TypeScript types from your C# code so you never have to keep the two in sync manually. This means your server and client logic stay tightly connected while you keep using the tools you already love. It is a brilliant way to modernize.NET apps without adding complexity.

📰 https://news.ycombinator.com/item?id=49163754

🆔 @hackernewsgithubprojects

📦 kem-a/appmanager

AppManager makes Linux app installs as easy as Mac

AppManager brings the familiar macOS drag-and-drop installation experience to Linux, turning the messy process of handling AppImages into a simple click-and-release workflow. Instead of digging through command lines, you just double-click a file or drag it into the app window, and it automatically extracts the software, sets up desktop shortcuts, and handles cleanup. It even manages background updates and lets you keep multiple versions of the same app side by side without conflicts.

🆔 @hackernewsgithubprojects

📦 kenton-gmi/sakura-crossing

Sakura Crossing: A Cel-Shaded Anime World in 3D

An explorable Japanese suburban neighborhood built entirely in 3D but rendered to look exactly like a hand-painted anime background. This project uses no image files at all. Instead, it draws every line and sign at runtime using code. You can walk through a shopping street, a shrine, and a railway crossing on a small spherical planet. The scene is filled with details like vending machines and a train that loops endlessly. You can even ride an e-bike through the town. It is a fascinating look at how to make 3D graphics feel like a classic 2D cartoon without using any traditional sprites or textures.

🆔 @hackernewsgithubprojects

📦 veedstudio/open-edit

OpenEdit

OpenEdit is the video editor that lets you style clips with a coding agent instead of a timeline. You tell your assistant to add stylized subtitles, and it handles the heavy lifting. The tool pulls the audio, generates precise word-by-word timings, and picks from a library of pre-designed visual templates to match your video. It renders the final clip locally on your Apple Silicon Mac, burning in those custom captions without needing a browser. You can then ask for tweaks, like changing colors or font sizes, and it updates the design instantly. It is a clever way to automate social media clips without touching a mouse.

📰 https://news.ycombinator.com/item?id=49169818

🆔 @hackernewsgithubprojects