📦 arjun988/blender-skills

blender-skills

The specialized toolkit that injects professional 3D-art knowledge straight into your favorite AI coding assistant, blender-skills gives AI agents the context they need to act like seasoned game artists. By feeding structured reference files into tools like Cursor or Claude, it prevents AI from hallucinating weird workflows and instead guides it to build optimized 3D pipelines. This means your assistant can write precise Python scripts to automate complex rigging, handle game engine exports, or generate stylized shaders without messing up. It is the ultimate bridge for developers who want to control Blender using AI without needing a degree in 3D modeling.

🆔 @hackernewsgithubprojects

📦 gazprash/2d-to-3d-voxelizer

2d-to-3d-voxelizer

You can now turn flat, two-dimensional pixel art into textured 3D voxel models with a single click, skipping complex AI models entirely. A clever tool named 2d-to-3d-voxelizer takes your flat sprites, or even multi-sided sprite sheets, and uses pure math algorithms to extrude them into 3D objects while keeping your original art style perfectly intact. It features different generation modes, like a quad-view that builds a full three-sixty object from a horizontal sheet, and you can export the final result as a standard 3D file for Blender or your favorite game engine. Grab a sprite sheet and start breathing life into your flat designs.

🆔 @hackernewsgithubprojects

📦 yaoliliu/freefuse

FreeFuse

FreeFuse is the training-free framework that finally lets you blend multiple subject LoRAs in image generators without ugly feature conflicts. Normally, when you try to combine different customized characters or styles into a single scene, their features bleed together and ruin the image. Instead of retraining models or forcing you to draw complex spatial masks, this tool automatically handles everything at the token level during generation. It analyzes the model's inner focus to map out where each subject belongs, then cleanly locks each LoRA to its designated area. It is a brilliant way to create rich, multi-subject scenes using your favorite base models.

🆔 @hackernewsgithubprojects

📦 1catai/1cat-vllm

1cat-vllm

Older enterprise graphics cards do not have to gather dust in the corner while newer hardware gets all the attention. The 1cat-vllm project optimizes the popular vLLM engine specifically for the older Tesla V100 graphics card, breathing new life into these classic processors. By adapting modern model structures and introducing custom memory techniques, this framework allows developers to run massive, state-of-the-art language models with deep context windows on older hardware setups. It solves the massive compatibility gap that usually locks older chips out of modern AI workflows. If you have some older hardware lying around, this is your ticket back into the game.

🆔 @hackernewsgithubprojects

📦 rebel-root/omni-browser

Omni Browser

Take full control of your mobile web experience with a privacy-focused Android browser that lets you run real desktop-grade extensions. Built on the same powerful engine as Firefox, omni-browser brings a suite of built-in tools directly to your phone, including a smart media grabber that sniffs out video streams and a private encrypted locker to keep your files secure. What makes it incredibly cool is that it lets you run powerful extensions like uBlock Origin and custom script overrides to bypass website copy restrictions or block annoying search assistant summaries. It is the ultimate toolkit for anyone wanting a faster, cleaner, and deeply customizable mobile web.

🆔 @hackernewsgithubprojects

📦 ekwek1/soprano

Soprano Ultra Fast Text to Speech

An ultra-lightweight text-to-speech model can run up to two thousand times faster than real-time on a standard graphics card while using less than one gigabyte of memory. Soprano is a compact, eighty-million parameter voice synthesis system designed to run entirely on your own local hardware. It solves the massive resource drain of typical AI voice generators, bringing high-fidelity, thirty-two kilohertz audio generation to standard consumer computers and even laptops without needing giant cloud servers. You can stream speech with under fifteen milliseconds of latency or spin up a local interface instantly. It is a fantastic tool for developers wanting fast, local speech generation without the heavy hardware tax.

🆔 @hackernewsgithubprojects

📦 miaai-lab/sparkdash

Sparkdash

Sparkdash is the multi-unit monitoring dashboard that manages several high-performance AI machines in a single window without any config restarts. It connects directly to your hardware to stream live graphics processor speeds, system memory usage, and local language model performance right to your screen. The standout feature is its ability to automatically detect which model engine you are running—like llama-cpp or vLLM—and track your actual token speeds in real time across different ports. It makes keeping tabs on your AI hardware setups incredibly simple, giving you instant performance insights at a glance.

🆔 @hackernewsgithubprojects

📦 amoghmunikote/cmpunlocker

cmpunlocker

Cmpunlocker is the open-source patching tool that turns a restricted graphics card back into a powerhouse. NVIDIA originally designed the CMP 170HX mining card using their top-tier A100 silicon, but they artificially cut its performance and memory in the firmware. This project works by patching NVIDIA's open-source Linux kernel drivers to bypass those artificial limits on startup. It tricks the hardware into unlocking its full computational muscle and expanding the usable memory from a tiny eight gigabytes up to sixty-four gigabytes. It is a fascinating peek into how software can reclaim locked hardware potential, giving tinkerers access to high-end enterprise computing power on a budget.

📰 https://news.ycombinator.com/item?id=48963891

🆔 @hackernewsgithubprojects

📦 archistico/shadeofcolor2

ShadeOfColor

Hide sensitive files right inside the pixels of a normal PNG image so they look completely harmless to anyone snooping around. ShadeOfColor2 lets you convert any block of data, even restricted executable files, into a standard picture that you can safely email or upload to strict cloud services. When it is time to get your data back, you just run the tool in reverse to extract the exact original file, complete with an integrity check to make sure nothing got corrupted. It is a brilliant, lightweight way to sneak your files past rigid filters without raising any red flags.

🆔 @hackernewsgithubprojects

📦 saivineeth147/lora-speedrun

LoRA Speedrun

The fastest verified training run holds the record in lora-speedrun, a public leaderboard dedicated to finding out how quickly you can fine-tune a language model. It acts like a referee for fine-tuning methods by freezing the hardware on a single graphics card and locking the task to a specific math or question-answering goal. Instead of relying on hype, developers submit their training code to be tested three times with fresh seeds in a secure sandbox. This project settles arguments about which speedup tricks actually pay for themselves in real-world training time and which ones fail to replicate, turning a competitive leaderboard into a shared lab notebook of proven techniques.

📰 https://news.ycombinator.com/item?id=48974325

🆔 @hackernewsgithubprojects

📦 bevibing/socrates-skill

Socrates Skill

Turn your artificial intelligence assistant into a personal tutor that refuses to give you the answers. By installing socrates-skill, you add a teaching mode to your developer tools. When you ask it about a complex codebase, an architecture diagram, or a document, the system will never hand over a direct solution. Instead, it guides you to the answer yourself through five levels of targeted questioning, from clarifying your initial assumptions to presenting hypothetical scenarios. If you get stuck or demand a hint, it simply breaks the problem down into smaller pieces, ensuring you actually learn how your systems work.

📰 https://news.ycombinator.com/item?id=47657874

🆔 @hackernewsgithubprojects

📦 unicity-aos/aos-ce

aos-ce

Build and run intelligent software in an open, highly secure sandbox. With aos-ce, you can run a lightweight, community-driven operating system designed specifically for managing autonomous agents. Instead of letting programs run wild, this system isolates tasks into individual, secure sandboxes called capsules that communicate over a protected event bus. The most exciting part is its self-building tool, which actually teaches agents to inspect their own environment, identify missing features during real tasks, and safely write and compile new capsules to expand their own capabilities. It is a completely modular, inspectable playground that turns agents from simple chat programs into self-improving tools.

🆔 @hackernewsgithubprojects

📦 boona13/mykonos-island-voxels

Mykonos Island Voxels

mykonos-island-voxels is the browser-based world builder that wraps the sun-bleached look of a Mediterranean holiday into a zero-install creative toy. Instead of grinding for resources or stressing over high scores, you get to drop beautiful, hand-painted blocks onto a tiny grid and watch a picturesque Greek village assemble itself right in front of your eyes. It is loaded with over seventy-five gorgeous, tiny assets like whitewashed cottages, cobalt-blue domes, and flowering bougainvillea. Best of all, it runs on pure JavaScript with no modern bundler bloat, letting you sculpt your own private coastal paradise on any device.

🆔 @hackernewsgithubprojects

📦 disler/fusion-harness

Fusion Harness

Combine the unique strengths of different frontier models simultaneously instead of constantly trying to pick a single winner. This clever extension harness lets you run a two-model agentic engineering team right from your coding workspace. It orchestrates roles, letting a planning model act as an architect while a building model writes the code, merging their independent output into one unified result with clear attribution. It even features a self-correcting validation loop where an automated gatekeeper tests the work and feeds failures back to the builder until the code is proven to work. Stop compromising on a single AI brain and put a coordinated, self-testing team to work on your next...

📰 https://news.ycombinator.com/item?id=48981521

🆔 @hackernewsgithubprojects

📦 xuefeng-cvr/depthart

depthart

depthart is the lightweight visual computer model that finally brings high-quality depth sensing to compact, everyday hardware. It takes a standard flat photo and accurately calculates the distance of everything in the frame down to the exact meter. While massive artificial intelligence models usually require giant servers to run these calculations, this clever project shrinks the math down so it runs directly on normal desktop graphics cards and tiny, portable microcomputers. It is perfect for robotics, smart cameras, or anyone wanting to build real-time spatial awareness into a small device without melting their hardware.

🆔 @hackernewsgithubprojects

📦 openlanguagemodel/openlanguagemodel

OpenLanguageModel

Build and experiment with custom transformer architectures without the headache of writing training pipelines from scratch. When you want to design a new neural network, you are usually forced to choose between massive frameworks that hide all the math, or writing thousands of lines of boilerplate code just to handle multi-GPU training and data loading. This library sits right in the middle, giving you readable PyTorch-native blocks for attention, norms, and feedforward layers while quietly managing the complex training loops behind the scenes. It is perfect for testing new ideas or learning how these networks actually tick. Check it out and start designing your own models today.

🆔 @hackernewsgithubprojects

📦 alibaba-damo-academy/rynnbrain

RynnBrain

Researchers just trained a new brain for physical robots that lets them handle complex real-world tasks with incredible agility. Developed by Alibaba's DAMO Academy, rynnbrain is an open embodied foundation model designed to give robots a smarter, highly generalizable way to perceive and interact with the physical world. Instead of hardcoding specific movements, this model acts as a general-purpose brain that translates sensory data into direct physical actions. It helps robots adapt to completely new environments and handle objects they have never seen before, bringing us much closer to having helpful, multi-purpose robotic assistants in our everyday lives.

🆔 @hackernewsgithubprojects

📦 ayghri/i-have-adhd

i-have-adhd

i-have-adhd is the coding assistant plugin that finally stops AI models from burying your answers in corporate fluff. Instead of making you scroll through polite small talk, vague explanations, and friendly closers, this tool forces coding agents like Claude Code and Cursor to give you action-first, highly structured responses. It enforces strict rules like leading with immediate commands, numbering multi-step tasks, and capping list items. You get a direct route to your next move, with zero conversational filler. It is a brilliant way to keep your developer workflow fast and completely distraction-free.

📰 https://news.ycombinator.com/item?id=48184192

🆔 @hackernewsgithubprojects

📦 hyscaler/hycanvas

HyCanvas

Design stunning graphics, social posts, presentations, and even videos in your browser without ever worrying about monthly subscriptions, paywalls, or forced watermarks. This project gives you a completely free, self-hostable design platform that runs entirely on your own terms. Instead of relying on closed ecosystems, you can run this whole suite from a single, lightweight binary on your own server or desktop, keeping every template and export completely private. It even supports a bring-your-own-key setup for integrating your favorite artificial intelligence tools securely. Self-host it today to reclaim absolute control over your digital creative studio.

📰 https://news.ycombinator.com/item?id=48988317

🆔 @hackernewsgithubprojects

📦 zli12321/lhtb

Long Horizon Terminal Bench

Test how well artificial intelligence actually handles real, grueling work by dropping it into a terminal for hundreds of steps. Instead of letting agents write one quick script and brag about it, lhtb forces them to operate in a live container where their work is graded by hidden, objective verifiers. They cannot just say they are finished; they have to keep working and adapting to feedback until they truly pass or run out of time. Even the best models only solve about a quarter of these tasks, and half are completely unsolved. It is the ultimate reality check for AI agents.

🆔 @hackernewsgithubprojects

📦 w3cj/how-llms-work

how-llms-work

You can train a decoder-only GPT transformer entirely from scratch in your web browser with zero machine learning libraries. An interactive web application called how-llms-work runs a complete educational pipeline locally using raw TypeScript. Instead of hiding the math inside complex frameworks, it implements every single step of modern language models by hand, from simple pattern matching and backpropagation to subword tokenizers and word embeddings. You can inspect active neural network layers, watch word vectors cluster together in real time, and trigger multi-threaded training operations inside a clean interface. It makes the actual mechanics behind generative artificial intelligence transparent and accessible.

🆔 @hackernewsgithubprojects