298 subscribers
14K videos
14.6K links
Download Telegram
This media is not supported in your browser
VIEW IN TELEGRAM
πŸ“¦ rjmahesh/ui-redline-walkthrough

The UI Redline Walkthrough

The UI Redline Walkthrough is the video tool that makes interface changes feel hand-drawn and real. Instead of just showing two static screenshots, it creates a film where red ink circles specific problems on the screen and writes short notes. Then the interface actually morphs into the fix, moving buttons and panels exactly like the real thing. It ends on the clean redesign, exporting a high-quality video file you can share anywhere. It turns a confusing before-and-after comparison into a clear, visual story of what changed. If you want to show off your design updates, this is the cool way to do it.

πŸ†” @hackernewsgithubprojects
This media is not supported in your browser
VIEW IN TELEGRAM
πŸ“¦ windshiftapp/core

Core: The lightweight Jira alternative

Take full control of your team’s workflow without the heavy footprint of Jira. This open-source tool packs boards, sprints, and a built-in knowledge base into a single, self-hosted application that runs on your own hardware. The standout feature here is that it actually connects coding agents directly to your work items, automatically linking pull requests to specific tasks so the code matches the plan. It handles the complexity of professional project tracking while staying simple enough to run from a single file. Your data stays on your servers, and you avoid expensive per-user licensing. It is a practical, powerful way to manage serious development work on your own terms.

πŸ“° https://news.ycombinator.com/item?id=50547381

πŸ†” @hackernewsgithubprojects
This media is not supported in your browser
VIEW IN TELEGRAM
πŸ“¦ hthoene/paperpaw

PaperPaw: The E-Reader Built from Text Files

Customize your e-ink reader without writing a single line of code. PaperPaw is an open-source firmware for small e-readers where every feature, from the home screen to file formats, is a pluggable module. The standout capability is that you can build a new plugin using only plain text files in a folder on the device. Want larger text or a new gesture? Just copy a few simple text files onto the reader, restart it, and the change is live. No compiling, no firmware flashing, no account, no ads. It runs on devices like the Xteink X4 and Android e-readers, keeping everything local and fully under your control.

πŸ†” @hackernewsgithubprojects
This media is not supported in your browser
VIEW IN TELEGRAM
πŸ“¦ tokenrhythm/infrarsi

InfraRSI: The AI That Optimizes Your AI Training

InfraRSI is the autonomous optimizer that makes your heavy machine learning jobs run significantly faster. If you are training large language models on a cluster of GPUs, you already know that squeezing every last bit of performance out of the hardware is a headache. This project takes the code you already have and uses artificial intelligence to hunt down specific bottlenecks, like slow data loading or inefficient memory use. It then tweaks the settings and patches the code to fix them. The real magic is that it tests every change to make sure your model still trains correctly before you accept it.

πŸ†” @hackernewsgithubprojects
This media is not supported in your browser
VIEW IN TELEGRAM
πŸ“¦ tubelabhq/youtube-research-skills

The AI Team Running Your YouTube Strategy

YouTube Research Skills is the AI toolkit that turns your chat assistant into a full-time YouTube strategy team. Instead of guessing what will work, it digs through real data to find exactly which niches are making money and how to break into them. The standout feature is that it acts like a six-person agency, spotting outlier videos that are punching above their weight to reveal hidden market gaps. It gives you specific, cited proof for every idea, so you stop relying on gut feeling and start building a channel on verified trends.

πŸ†” @hackernewsgithubprojects
This media is not supported in your browser
VIEW IN TELEGRAM
πŸ“¦ gosu-x/multimodal-threat-dataset

Multimodal Threat Dataset

A threat dataset for multimodal networks contains exactly one file: a single text document that simply states its own name and purpose. There is no code, no data, and no actual threat information to analyze. This project serves as a placeholder for future research into how language and image models can be tested against complex attacks. While currently empty, it signals an effort to build standards for measuring these risks. For now, it is a blank slate rather than a tool, showing that the work is just beginning.

πŸ†” @hackernewsgithubprojects
This media is not supported in your browser
VIEW IN TELEGRAM
πŸ“¦ chriskearney/siliconforest

SiliconForest: Mac Apps on Open Source

SiliconForest is an open operating system running natively on an M1 MacBook Pro that executes unmodified Mac applications without a hypervisor. It solves the gap between Apple’s closed ecosystem and open source by booting directly on the hardware using the same Darwin kernel foundation as macOS. You get a real GPU-accelerated desktop where Firefox and native system tools run exactly as they do on a standard Mac. This is not a simulator; it is the actual machine talking to actual code. The takeaway is simple: you can now run genuine Mac software on a fully open, transparent system that you understand at a fundamental level.

πŸ“° https://news.ycombinator.com/item?id=50542257

πŸ†” @hackernewsgithubprojects
This media is not supported in your browser
VIEW IN TELEGRAM
πŸ“¦ yyfz/pi4

Pi4: A Foundation Model for 4D Scene Geometry

Pi4 is the foundation model that reconstructs a full 3D scene and tracks its motion in a single forward pass. It solves the problem of needing separate tools for geometry and motion by processing thousands of video frames together, without any camera calibration, to build a shared spatial representation. The surprising strength is handling long sequences like a minute-long figure skating clip in one go, outputting precise 3D points, camera paths, and moving object masks simultaneously. You can even spin up a browser viewer to orbit the reconstruction in real time.

πŸ†” @hackernewsgithubprojects
This media is not supported in your browser
VIEW IN TELEGRAM
πŸ“¦ sirioberati/ai-filmmaking-camera-control-seedance-2-5

Control AI Camera Moves with Code

ai-filmmaking-camera-control-seedance-2-5 is the filmmaking tool that lets you script exact camera moves in code instead of guessing with prompts. It solves the problem of unpredictable AI video by giving you a 3D map to follow. You write a simple path for the camera, like a dolly zoom or an orbit, and the AI video model copies that motion perfectly. The clever trick is using striped pills as stand-ins for people. This stops the AI from copying body movements it shouldn't have, while keeping the camera path precise. It’s like handing the AI a puppet guide for the lens, not the actors.

πŸ†” @hackernewsgithubprojects
Media is too big
VIEW IN TELEGRAM
πŸ“¦ yeufonic/yeufonic

Yeufonic: Make Real Songs on Your PC

Yeufonic lets you generate full songs with editable sheet music entirely on your own computer. It is a local web app that turns your ideas or recordings into actual music, solving the problem of needing expensive studios or cloud services to make real songs. The most interesting part is that you can cover your own recordings or write new pieces from a simple prompt, then edit the score like a pro before exporting the final audio. Everything runs locally on your machine, meaning no cloud dependencies and full privacy. If you want to make music without leaving your desk, this tool gives you the complete toolkit right in your browser.

πŸ†” @hackernewsgithubprojects
This media is not supported in your browser
VIEW IN TELEGRAM
πŸ“¦ shoestring-dev/strixfold

Run DeepSeek on Two Cheap AMD Boxes

Run a massive AI model on two affordable AMD computers without needing expensive NVIDIA hardware. Strixfold takes two Strix Halo machines and links them with standard USB4 cables to work together as one powerful system. It runs DeepSeek-V4.1-Flash, delivering about seventy-two percent of the speed of high-end NVIDIA rigs while producing identical results. The coolest part is how it splits the heavy math between both boxes over simple network cables, proving you don't need a data center to get serious performance. If you have spare hardware sitting around, this is your ticket to building a serious local AI setup. Check out the repo to see how it’s done.

πŸ†” @hackernewsgithubprojects
This media is not supported in your browser
VIEW IN TELEGRAM
πŸ“¦ jackson-cmd/claude-multi-desktop

Claude Multi Desktop: Chat Across Your Accounts

Stop juggling your personal and work Claude accounts on Mac. With Claude Multi Desktop, you can run both side by side, each with its own sign-in, while sharing one single coding setup. The coolest part? Your sessions can literally talk to each other. You can ask your work session to check a list your personal session is managing, and it just happens. It is one simple script, no heavy software, and no mess. If you use two accounts, this is the clean way to stop the chaos and let your bots collaborate.

πŸ“° https://news.ycombinator.com/item?id=50520258

πŸ†” @hackernewsgithubprojects
This media is not supported in your browser
VIEW IN TELEGRAM
πŸ“¦ kmic-68/llama.cpp

Llama dot cpp

Run large language models on your own computer with Llama dot cpp. This open-source project lets anyone execute artificial intelligence locally without needing a cloud server or massive new hardware. It is particularly interesting because it has been specifically tuned to make older graphics cards run these models much faster than standard versions, pushing performance nearly twice as high while keeping the math incredibly accurate. By handling the heavy lifting right on your machine, it gives you full control over your data and processing speed. You get a powerful, private AI assistant that works reliably on equipment you already own.

πŸ“° https://news.ycombinator.com/item?id=50146283

πŸ†” @hackernewsgithubprojects
Media is too big
VIEW IN TELEGRAM
πŸ“¦ yang-mills-group/openai-ym-20261028

OpenAI Proves Yang-Mills with Lean

Stop guessing and start proving. This project takes the legendary Yang-Mills problem and solves it using the Lean theorem prover, turning massive mathematical concepts into computer-checked logic. It doesn’t just claim a result; it builds a rigorous bridge from lattice calculations to physical reality, ensuring every step is mathematically perfect. For the first time, we have a machine-verified proof of this cornerstone of quantum physics, eliminating human error from the process. If you’ve ever wondered how we can be absolutely certain about the math behind the universe, this is your answer. It’s not just code; it’s certainty you can run.

πŸ†” @hackernewsgithubprojects
This media is not supported in your browser
VIEW IN TELEGRAM
πŸ“¦ agentic-learning-ai-lab/deleting-weights

The deleting-weights

Teach a large language model new skills without touching nearly all of its weights. The deleting-weights project shows that zeroing out a tiny fraction of a model's internal connections can make it just as capable as a full, heavy update. This approach, called Deletion Tuning, solves a massive headache for developers who need to add features without losing what the model already knows. It keeps the original knowledge intact while the model learns new tasks, all while using a file size that is thousands of times smaller than standard updates. It is a clever way to make AI models smarter and leaner at the same time.

πŸ†” @hackernewsgithubprojects
This media is not supported in your browser
VIEW IN TELEGRAM
πŸ“¦ anionex/h3-image-mlx

H3 Image MLX: Local AI Art on Your Mac

Generate and edit stunning images directly on your Mac using the H3 Image MLX library. It takes a popular AI model and runs it entirely on your computer’s built-in graphics chip, so you don’t need a cloud subscription or a supercomputer. The coolest part is how it manages memory. Instead of keeping the whole massive model loaded in your random access memory, it reads the data in tiny, on-demand chunks and releases it instantly. This lets you create huge, high-resolution photos using only a few gigabytes of space, even while your browser and other apps are running.

πŸ†” @hackernewsgithubprojects
This media is not supported in your browser
VIEW IN TELEGRAM
πŸ“¦ budgetany/budget3d

Budget3D: One Model, Any Quality

Budget3D is the image-to-3D tool that lets you dial down quality without losing your mind. It takes a single photo and turns it into a spinning 3D object, but here is the cool part: you can tell it to use less computer power for the boring parts, like empty space, while keeping the details sharp where they matter. One model handles every setting, so you are not switching between different files for high or low res. It is like a volume knob for your 3D scene, letting you decide exactly how much detail you need.

πŸ†” @hackernewsgithubprojects
This media is not supported in your browser
VIEW IN TELEGRAM
πŸ“¦ nvlabs/dvsm

DVSM: Real-Time 3D From a Single Transformer

DVSM is the view synthesis model that renders entirely new 3D scenes using just a single Transformer network. It solves the problem of needing a separate, slow optimization step for every new environment by reconstructing a scene and painting its pixels in one continuous pass. This is particularly exciting because it matches the quality of traditional methods that take much longer to process. You feed it a handful of photos, and it immediately lets you explore that space from any angle in real-time. It is a fast, self-contained way to build immersive 3D environments from simple flat images. This tool makes complex scene generation accessible and immediate.

πŸ†” @hackernewsgithubprojects
Media is too big
VIEW IN TELEGRAM
πŸ“¦ vincalkr/almond

Running Frontier AI on a Gaming GPU

Almond runs massive artificial intelligence models directly on standard consumer graphics cards. It handles complex language engines on a single or dual AMD Radeon setup. The software squeezes these heavy computational tasks into your existing video memory. It bypasses the need for expensive enterprise server hardware. This open source project brings high-end local inference to everyday creators. You can test powerful neural networks without renting cloud time. It democratizes access to advanced language generation tools. The engine is optimized specifically for consumer silicon performance. This is how you run next-gen AI on your desktop. No data center required, just your graphics card. This is Almond.

πŸ†” @hackernewsgithubprojects
Media is too big
VIEW IN TELEGRAM
πŸ“¦ spearsheep/anima-brain

Anima Brain: AI that actually remembers you

Your AI companion forgets everything the moment you close the tab. Anima Brain fixes that by giving it a local, private memory that tracks your facts, feelings, and even how your relationship evolves over time. Instead of just storing text, it connects past conversations to your current mood, so your chat partner feels consistent and caring rather than robotic. It runs entirely on your computer using a simple database, meaning your conversations stay yours and don't get sent to a distant server. If you want your AI friend to actually know you, this is the tool that makes it happen without the cloud headache.

πŸ†” @hackernewsgithubprojects