📦 gronke/ui-components

One Component, Two Screens: TUI and Web in Rust

ui-components lets you write a single user interface in Rust and watch it run as a full web page in your browser and as a crisp terminal app at the same time. You define the layout and logic once, then the project builds two identical interfaces that share the exact same keyboard shortcuts, properties, and events. A real date picker looks like a standard web form on the desktop but transforms into a neat box widget inside a terminal window. It even works without JavaScript in the terminal, using real CSS flexbox for layout. If you change the data in one view, the other updates instantly.

🆔 @hackernewsgithubprojects

📦 willjim/kiri-maker

Kiri Maker: Hand-Controlled 3D Splatting

Kiri Maker lets you control 3D models with your hands. It is a browser tool that takes public 3D scans from apps like Polycam and turns them into sparkling particle clouds. You can watch the scene dissolve into digital dust and then reassemble with a wave. There is a handy dice button for testing without your own links. You can also pick from sixteen cinematic camera paths or draw your own flight route. The real magic is the gesture control. Hold up a palm to scatter the particles. Clench a fist to gather them back together. Point to orbit the view. Two fists moving apart scales the model.

🆔 @hackernewsgithubprojects

📦 circuitlord/bigwalkvrinstaller

Big Walk VR Installer

Big Walk VR Installer lets you add full virtual reality support to the puzzle game Big Walk with a single click. Instead of manually downloading files, this tool automatically finds the game on your computer, installs the necessary mod loader, and sets up the VR environment for you. The coolest part is how it handles multiplayer. If you join a game without VR, you still see other players' real hand movements, even if they are flatscreen. It keeps everything tidy by remembering exactly what it installed, so updates never break your settings. It makes a complex mod setup feel completely simple.

🆔 @hackernewsgithubprojects

📦 ychangqing/igraph

IGraph: Turn Code Into a Knowledge Graph

IGraph is the code analyzer that turns messy repositories into searchable knowledge graphs. Most tools just index files, but this one builds a living map of every function and variable, then links them with a smart AI engine so you can ask things like where authentication happens and get the exact answer. It even lets you attach design docs and database schemas to see how they connect to the actual code. It works with Cursor and Claude Code out of the box. Stop guessing where things live and start seeing the whole picture. Just one line of code to get started.

🆔 @hackernewsgithubprojects

📦 nicolab28/comfyui-clipproj

Cut Video Model Size by 70 Percent

Comfyui Clipproj lets you swap a massive text encoder for a tiny one, shaving roughly eleven gigabytes off your video model without losing quality. It works by training a simple linear map that translates small model outputs into the large model space, letting you run MiniMax H3 on cards with barely eight gigs of memory. The creator built this as a curiosity project to understand how diffusion models use text, and it actually produces good video despite being a proof of concept. It uses ridge regression instead of deep training, making the setup surprisingly simple and lightweight. This is a fascinating experiment in squeezing performance out of smaller hardware.

🆔 @hackernewsgithubprojects

📦 airweave-ai/error-monitoring-agent

Error Monitoring Agent

Turn noisy error logs into smart, actionable alerts with the Error Monitoring Agent, a tool that automatically groups similar crashes and hunts down the real reason they happened. Instead of drowning you in duplicate notifications, it clusters issues by root cause and then searches your entire codebase, past tickets, and Slack chats to find exactly who worked on this before or which code file is breaking. This means you get clear context about whether something is a new bug or just a known glitch, stopping alert fatigue instantly.

🆔 @hackernewsgithubprojects

📦 mit-zardini-lab/pyncd

Pyncd: Algebraic Deep Learning Diagrams

Pyncd lets you build complex deep learning models like transformers or DeepSeek-V3 using simple Python algebra. Instead of writing hundreds of lines of low-level code, you compose mathematical operators with standard symbols, and the library automatically handles axis alignment and broadcasting. You can then visualize these models as detailed circuit diagrams or compile them directly into PyTorch modules for execution. This bridges the gap between abstract category theory and practical machine learning implementation, making it easier to reason about model structure. It is a neat tool for developers who want to see the mathematical skeleton of their neural networks clearly.

🆔 @hackernewsgithubprojects

📦 algonorhythm/flare

Flare: The IDE That Maps Your Code Like a City Map

Flare turns your codebase into a living map so you can instantly see how every file connects. Instead of staring at a boring file list, you watch a graph update as an AI agent works, with colors showing which files are risky or barely tested. The coolest part is the Review Cockpit, which groups changes into clear batches so you can spot when an agent breaks something and revert it instantly. It feels less like writing code and more like navigating a city where every street and building has a story.

🆔 @hackernewsgithubprojects

📦 suxrobgm/jobpilot

JobPilot: The AI Agent That Applies to Jobs for You

JobPilot is the AI job application agent that actually does the clicking and filling for you, running locally on your own machine so your data stays private. Instead of manually clicking through dozens of job boards, this tool uses your personal Claude or Codex subscription to search for roles, tailor your resume, and even draft messages to recruiters. The coolest part is that it runs in a real browser window right on your laptop, meaning it can beat those pesky CAPTCHAs and fill out forms exactly like a human would. You keep full control because the heavy lifting happens on your own subscription, not theirs.

🆔 @hackernewsgithubprojects

📦 coko7/latuicon

latuicon: The Terminal Emoji Picker

latuicon is the terminal interface that lets you pick emojis, kaomoji, and symbols without leaving your command line. It solves the headache of hunting through endless menus just to find one cute little face for your scripts. You simply type a few letters to fuzzy search, hit enter to grab the icon, and it prints it straight to your clipboard or variable. It even handles typos gracefully, so you don't need perfect spelling to find that thumbs-up emoji. The best part is how cleanly it fits into your workflow, letting you capture the result with a single line of code.

🆔 @hackernewsgithubprojects

📦 dominikmartn/progressiveblurheader

ProgressiveBlurHeader: Apple-style blur effect

ProgressiveBlurHeader is the SwiftUI component that finally nails the tricky Apple Music style blur without the usual headaches. Building sticky headers with variable blur is notoriously hard because SwiftUI lacks a native API, and most workarounds end up with ugly hard edges or flat, uniform blurs that clip your content. This tool solves that by stacking a scrolling view, a custom blur layer, and your header in a simple three-layer setup that keeps everything visible and smooth. It uses the same internal blur engine found in the App Store, giving you full control over intensity and tinting while automatically adapting to light and dark modes.

🆔 @hackernewsgithubprojects

📦 modeltc/minimax-h3-turbo

Minimax-H3 Turbo: Generate Video in 4 Steps

Minimax-H3 Turbo makes high-quality video generation incredibly fast by cutting the required processing steps down to just four. This project distills the original heavy model into lightweight versions that run significantly quicker on your hardware. You can use these checkpoints with standard coding libraries or visual workflow tools, allowing you to create videos from text prompts or existing images with synchronized audio. The repository provides ready-to-use setups for both technical users who prefer code and creators who like visual interfaces. It handles various video resolutions and keeps the original style intact while speeding up the creation process.

🆔 @hackernewsgithubprojects

📦 duckyshell/comfyui-minimaxh3-prompt-writer

ComfyUI MiniMax H3 Prompt Writer

ComfyUI MiniMax H3 Prompt Writer lets you generate complex video prompts by mixing plain language with photos and clips. You describe your idea, add reference images or video clips, and the tool uses a local AI model to craft a detailed prompt for the MiniMax H3 video generator. It handles all the technical syntax so you just focus on the creative brief. The setup is flexible, supporting local models via Ollama or direct GGUF files, or remote APIs if you prefer. You get an editable result that you can refine or copy directly into your workflow.

🆔 @hackernewsgithubprojects

📦 shmimel/bee-write-back

Bee Write Back: The Dedicated Journaling Device

Write your thoughts into a private journal without letting your whole past come rushing back to haunt you. Bee Write Back is a clever Raspberry Pi device built around the WriterDeck form factor, designed specifically for nightly journaling. You open the app, answer a random prompt or just freewrite, and then lock it away until the next day. The beauty is in the separation: unlike a digital notebook that shows you every old entry, this system keeps your reflections fresh and focused on the present moment.

🆔 @hackernewsgithubprojects

📦 tonyd2wild/deepseek-v4-flash-dgx-spark

DeepSeek-V4-Flash on dual NVIDIA DGX Spark

Setting up multi-node serving on dual NVIDIA DGX Spark boxes usually ends in tears, with broken RDMA handshakes, cold-start timeouts, and silent driver bottlenecks. The deepseek-v4-flash-dgx-spark repository delivers a battle-tested recipe to run the DeepSeek-V4-Flash mixture-of-experts model split across two Spark nodes using vLLM. It orchestrates a tensor-parallel configuration over a two-hundred gigabit Ethernet RoCE link, utilizing FP8 quantization and multi-token prediction speculative decoding. The project includes a production-ready Docker image with native kernels, launch scripts, and solutions for the dreaded GID index mismatch. Sync your firmware, spin up the worker, and enjoy blazingly fast, long-context local serving.

🆔 @hackernewsgithubprojects

📦 carasibana/comfyui-h3-facerefine

Fix Blurry Faces in MiniMax H3 Video

MiniMax H3 struggles to render faces clearly when characters appear small in the frame. Comfyui-h3-facerefine solves this by automatically detecting each face, cropping it so it fills the screen, and letting the video model regenerate the details before stitching the result back into the original clip. This approach works because the model gets a close-up view instead of a distant blur, producing sharper expressions without changing the overall shot. The process runs frame by frame, tracking movement smoothly to keep the video stable. It is a practical way to fix a common generation flaw, giving you high-quality results even when the subject is far away.

🆔 @hackernewsgithubprojects

📦 igniteui/igniteui-mcp-testbed

Test if AI Actually Uses Your Tools

See if your AI agent actually uses the tools you gave it by running the Ignite UI MCP Testbed. It spins up a fresh, isolated container, scaffolds a web app, and hands it to an AI coding agent with specific MCP servers and skills enabled. The real magic is the History tab, which tracks exactly which tools the AI called, which skills it loaded, and—crucially—what it ignored. You can instantly see if your agent is ignoring your custom APIs or just ignoring everything. It solves the mystery of AI reliability by proving whether your setup actually works, not just whether the code compiles.

🆔 @hackernewsgithubprojects

📦 matlowai/comfyui-mainodes

ComfyUI MAINodes Fixes Smearing in Fast Motion

ComfyUI MAINodes adds two clever nodes to MiniMax-H3: a Motion Lab that stops fast-motion smearing by temporarily slowing down only the chaotic parts of a clip, and a Contact-Sheet node that generates five simultaneous views of a subject from one reference image. The Motion Lab works by identifying where the model struggles to keep up, stretching out those specific frames so the AI gets extra time to draw clear poses, and then snapping them back to normal speed afterward. It uses a smart oracle to spot the messy spots without needing extra training, leaving the stable background untouched.

🆔 @hackernewsgithubprojects

📦 donkeyusecorp/donkey

Donkey

Donkey is the local-first video editor that finally keeps your media safe on your own machine. Unlike cloud-heavy editors that upload your footage, Donkey runs its heavy lifting right inside your Mac. You get a full timeline with multi-track editing, captions, and effects, all processed locally so your files never leave your disk. The real magic happens with its AI assistant. You simply describe the shot you want, like a 1920s comic chase scene, and the app generates the video, voice, and music for you.

🆔 @hackernewsgithubprojects

📦 mesh2motion/mesh2motion-desktop

Mesh2Motion Desktop

Mesh2Motion Desktop is the offline desktop app that brings web animation tools to your local machine without the internet. Right now, it is a work in progress using the Tauri framework, which keeps the app tiny by packing Rust code into small binaries instead of heavy web standards. The core idea is wrapping the existing Mesh2Motion web experience into a native window, letting you browse and download animations directly from your file system. It currently saves files to downloads and includes developer tools, so it is still in early testing.

🆔 @hackernewsgithubprojects

📦 nathanw1014/strix-halo-llamacpp

Strix Halo llama.cpp: Blazing Fast Local AI

Strix Halo llama.cpp delivers massive speedups for AMD’s new Strix Halo graphics chips by bundling specialized fixes directly into a ready-to-run package. Most developers struggle with slow performance when running complex AI models on this specific hardware, but this project solves that by optimizing how the chip processes long contexts and quantized data. Instead of fighting with complicated build tools, you get a self-contained folder that includes the necessary GPU drivers, so you can start running large language models immediately. The result is a dramatic jump in processing speed, making it feasible to run heavy AI models locally without needing expensive server-grade hardware.

🆔 @hackernewsgithubprojects