📦 miaai-lab/qwen3.6-27b-nvfp4-vllm

Qwen 3.6 27B NVFP4 vLLM

Spin up a high-performance, private AI server on your own hardware in just minutes. The project qwen3.6-27b-nvfp4-vllm gives you a complete, ready-to-run setup for hosting a highly optimized twenty-seven billion parameter language model. It bundles everything into a simple Docker container, complete with automated start and stop scripts, so you do not have to struggle with complex configuration. Because the model is compressed to a highly efficient four-bit format, it fits comfortably on standard forty-eight gigabyte graphics cards while still supporting massive text histories and multi-modal image inputs. Grab the repository, run the launch script, and enjoy your own local, blazing-fast chat assistant.

🆔 @hackernewsgithubprojects

📦 miaai-lab/qwen3.6-27b-nvfp4-vllm

Qwen 3.6 27B NVFP4 vLLM

Spin up a high-performance, private AI server on your own hardware in just minutes. The project qwen3.6-27b-nvfp4-vllm gives you a complete, ready-to-run setup for hosting a highly optimized twenty-seven billion parameter language model. It bundles everything into a simple Docker container, complete with automated start and stop scripts, so you do not have to struggle with complex configuration. Because the model is compressed to a highly efficient four-bit format, it fits comfortably on standard forty-eight gigabyte graphics cards while still supporting massive text histories and multi-modal image inputs. Grab the repository, run the launch script, and enjoy your own local, blazing-fast chat assistant.

🆔 @hackernewsgithubprojects

📦 img2threejs/img2threejs

img2threejs

img2threejs is the image to three d generator that turns a single reference photo into a clean, animation ready, code only three js model. Instead of giving you a massive, messy mesh file or a blurry photo projection, this clever tool writes a lightweight typescript factory file that builds the object using clean geometric primitives and procedural shaders. It works in structured, quality gated passes, checking things like joint attachments and realistic materials to make sure the parts actually line up and move naturally. It is incredibly token efficient because clever python scripts handle the heavy lifting while the artificial intelligence only steps in to judge the visual quality. If...

🆔 @hackernewsgithubprojects

📦 songguoxs/seedance-prompt-skill

Seedance Prompt Skill

Design cinematic, production-ready video prompts for ByteDance's video generator right from your command line. The seedance-prompt-skill is a specialized custom tool for Claude Code that turns the AI into a professional prompt engineer. You simply describe your creative concept in natural language, and the tool outputs structured, multi-modal Chinese prompts tailored for the generator's advanced capabilities, including camera motion replication, storyboard continuation, and video editing. To get started, clone the repository, copy the skill folder into your Claude directory, and launch your workspace. Take control of your AI filmmaking process and bring your visual ideas to life instantly.

🆔 @hackernewsgithubprojects

📦 grp06/useful-codex-skills

useful-codex-skills

You can now completely automate how your coding assistant plans, pressure-tests, and executes complex software refactors. useful-codex-skills is a clever collection of modular workflows that teach your AI assistant to act like a seasoned software architect. Instead of writing messy, uncoordinated code, it uses these tools to run structured decision-tree interviews, turn vague tasks into highly detailed execution plans based on classic software design principles, and even publish clear explanations of code changes directly to your connected Notion workspace. It is a brilliant way to ensure your development agents always produce simple, well-tested, and predictable code on the first try.

🆔 @hackernewsgithubprojects

📦 laskco/mpvnova

mpvnova

Your Android TV can finally play high quality videos smoothly without stuttering or compatibility issues. That is because mpvnova is a media player built specifically for television screens, bringing the legendary power of mpv directly to your living room. Instead of struggling with sluggish interfaces designed for touchscreens, this app gives you a custom television user interface, advanced audio delay controls, and specialized decoder tools to handle tough video files. It even includes clever workarounds for popular streaming hardware, like fallback options for Nvidia Shield devices, ensuring your movies play perfectly every time. It is the ultimate way to make your smart TV actually smart.

📰 https://news.ycombinator.com/item?id=48085168

🆔 @hackernewsgithubprojects

📦 ancientjames/scanwheel

Scanwheel mechanical television

Build your own working mechanical television using 3D printed parts and a Raspberry Pi Pico. Scanwheel is a drum-style mechanical TV featuring five side-by-side viewing windows that share a single spinning drum. Tiny pinholes in the drum act as physical pixels, while separate, synchronized light-emitting diodes recreate images behind each window. By using the Pico's specialized hardware timers, the system achieves incredibly high horizontal resolution, making the twenty-line display look surprisingly detailed and sharp. It is a brilliant way to bring early twentieth-century television technology into the modern era using simple, accessible maker electronics.

📰 https://news.ycombinator.com/item?id=49047707

🆔 @hackernewsgithubprojects

📦 ahapbean/xhc

xhc

Scale artificial intelligence models wider and push their intelligence further by unlocking massive parallel pathways. This is made possible by xhc, a breakthrough design that expands the standard single connections inside neural networks into sixteen parallel streams. Usually, adding more streams slows computers to a crawl, but this method reads all sixteen streams while only updating four active ones at a time. It also uses smart temporal features to make sure the information stays rich and diverse. This clever shortcut delivers massive performance boosts with barely any extra computing cost. It is a brilliant way to make big models much smarter without needing more hardware.

🆔 @hackernewsgithubprojects

📦 infrarecon7/ir275k

IR275K

Standard algorithms struggle to sharpen infrared video because typical training data hides real-world flaws like weak heat contrast and shaky camera motion. The new benchmark dataset called ir275k fixes this by curating over two hundred and seventy-five thousand frames of remote sensing infrared footage into strict, sequence-level tests. Instead of letting algorithms cheat with easy random frames, it challenges them with heavy noise, low-contrast landscapes, and fast motion across consecutive frames. It is a fantastic tool for developers wanting to build thermal imaging models that actually hold up in harsh, real-world conditions. Check it out if you want to push your computer vision models to the limit.

🆔 @hackernewsgithubprojects

📦 0xwilliamortiz/agents-council

Agent Council

You can now run a local panel of diverse artificial intelligence assistants right from your command line without paying any extra application programming interface fees. An open source tool called agents council acts as a plugin for your terminal assistant to orchestrate multiple command line tools simultaneously. When you ask a complex technical question, the tool queries different platforms like Claude Code, Codex, and Gemini in parallel. It gathers their individual perspectives, displays their unique feedback, and has a main host agent synthesize everything into one clear, balanced recommendation. It is a brilliant way to get a second opinion on your code using the developer tools you already have installed.

🆔 @hackernewsgithubprojects

📦 ringhyacinth/meow-generator

Meow Generator

You can now procedurally generate your own custom 3D kitten right in your web browser. Meow-generator is a local-first creative tool that uses math to craft unique, highly customizable felines. Instead of loading heavy pre-made 3D models, it builds everything from scratch, letting you adjust body fluff, pupil size, eye colors, and even heterochromia. You can interact with your digital kitten by tossing physics-based toys, lifting them by the neck, and poking their soft cheeks. It even features customizable weather, playable animations, and collectible cards, making it the perfect creative sandbox for any cat lover.

🆔 @hackernewsgithubprojects

📦 sterzhang/awesome-progress-models

awesome-progress-models

Help robots understand whether they are actually getting closer to a goal, standing completely still, or making a giant mess and undoing all of their hard work. The awesome-progress-models repository is a curated map of research papers, methods, and benchmarks dedicated entirely to progress reward modeling for robotic learning. Instead of just checking if a robot finished a task at the very end, these models evaluate its step-by-step progress in real time using vision, language, and temporal data. It is a fantastic, highly organized resource for anyone trying to build smarter robots that can learn from their own mid-task mistakes.

🆔 @hackernewsgithubprojects

📦 7-e1even/learn-agent

learn-agent

learn-agent is the AI development guide that finally shows you how to make autonomous software reliable instead of just functional. Created during the development of a real desktop agent, this project breaks down the exact internal mechanics of coding tools like Claude Code and simplifies them into zero-dependency, single-file Node programs you can run instantly. Instead of guessing from theory, it tackles the messy, real-world edge cases like handling budget overflows, recovering from sudden user interruptions, and keeping the model from running dangerous terminal commands. It is a masterclass in building AI systems that actually survive the real world.

🆔 @hackernewsgithubprojects

📦 victortaelin/optmem

OptMem

Give your artificial intelligence agents a permanent memory that outlives any session, model update, or vendor change. OptMem is a simple, lightweight tool that lets coding assistants and autonomous agents build a structured history of their work without cluttering their prompt window. By adding a short instruction block to your agent's system files, it learns to log key decisions and search past notes using a tiny Python script. Instead of drowning in thousands of old chat tokens, the agent summarizes its memory into a clean binary tree, allowing it to recall critical facts instantly while keeping operating costs low.

📰 https://news.ycombinator.com/item?id=49061311

🆔 @hackernewsgithubprojects

📦 swellweb/reame

Reame

Run useful language models on cheap, free-tier ARM servers instead of renting expensive graphics cards. Reame is a lightweight inference server built specifically for the CPUs you already own, designed around the idea that a processor should never compute the same thing twice. By combining an on-disk prompt cache that survives restarts with smart speculative decoding, it slashes response times on repetitive tasks. In real-world testing, simply reformatting a document into a clean fact sheet let a tiny model run six times faster and achieve a perfect accuracy score. It is the perfect setup for private, budget-friendly artificial intelligence features.

📰 https://news.ycombinator.com/item?id=48873417

🆔 @hackernewsgithubprojects

📦 dmae97/omk

omk

omk is the provider-neutral control plane that finally brings reliable order to AI coding assistants. Instead of letting AI models run wild with no accountability, this tool acts as a structured coordinator. It takes your goals and maps them into safe, parallel lanes so multiple tasks can run at once without conflicts. Best of all, it features evidence gates that block completions until your changes actually pass real verification. You can easily swap between models like OpenAI and Anthropic without changing your workflow. It is the ultimate way to keep your AI developers honest, organized, and reliable.

🆔 @hackernewsgithubprojects

📦 aashishh15/lexicon

Lexicon

Lexicon is the distraction-free desktop editor that finally acts as a completely offline, private alternative to cloud writing assistants like Grammarly. Running entirely on your Windows or Mac computer without sending a single byte of your text to the cloud, it gives you a clean workspace for rich text, markdown, and LaTeX formulas. It bundles its own local grammar checker to catch typos and punctuation issues instantly, alongside an optional local AI assistant that helps you rewrite sentences, change your writing tone, or summarize long drafts. Because everything runs on your own hardware, you get full privacy, no subscription fees, and an outstanding writing environment that works perfectly even when...

🆔 @hackernewsgithubprojects

📦 buage/selfhostlist

SelfHostList

Discover awesome, independent software you can run right on your own home hardware. selfhostlist is a curated directory designed to help you break free from big cloud subscriptions by finding open source alternatives you control completely. It indexes dozens of self-hosted applications and services, featuring an easy search bar and tag filtering to help you pin down exactly what you need. Whether you want to host your own media server, a private cloud, or personal productivity tools, this list makes finding the right software simple. Check out the project to start hosting your own digital life today.

🆔 @hackernewsgithubprojects

📦 blakeem/navidrome-mcp

Navidrome MCP

Control your personal music server using nothing but natural language. By connecting to your existing Navidrome library, this clever tool lets you chat with your music database to build smart playlists, discover missing albums, and fetch synced lyrics on the fly. The absolute best part is the local audio playback through mpv, which turns your computer into a hands-free music player. You can just ask to queue up a mix of your favorite nineties rock songs, and it instantly plays straight through your speakers, syncing everything back to your server. It is a brilliant way to bring a classic self-hosted music setup into the modern era.

🆔 @hackernewsgithubprojects

📦 lyallcooper/justif

Justif

Justif is the JavaScript library that brings beautiful, publication-grade text justification directly to the web. Instead of the browser's standard line-by-line justification, which often leaves awkward gaps and uneven spacing, this library uses a sophisticated paragraph-wide algorithm to evaluate layout as a whole. It introduces polished typography features like hanging punctuation that extends into the margins, automatic hyphenation across multiple languages, and subtle adjustments to variable font widths to create incredibly clean, balanced text blocks. It works as a progressive enhancement, leaving unsupported browsers or styles completely untouched while upgrading compatible text into gorgeous, print-quality prose.

🆔 @hackernewsgithubprojects