📦 sergiiob/intel-arc-pro-b70-inference-cookbook

Battlemage GPU Hacks: 160 Tokens Per Second

A single Intel Arc Pro B70 graphics card is pushing a massive thirty-five billion parameter AI model at one hundred sixty tokens per second, which sounds impossible for a discrete GPU with that price tag. This open cookbook gives you the exact recipes, engine patches, and benchmark harnesses needed to get those numbers on Intel's Battlemage hardware. It solves the problem of opaque performance data by proving exactly how to configure inference engines for maximum speed on these specific chips. If you are looking to run large language models locally on Intel graphics, this is your definitive guide to making it actually fly.

🆔 @hackernewsgithubprojects

📦 xmanrui/dsh-im

Turn Any Chat App Into Your AI Agent

Turn any chat app into a direct line to your local AI agent. DSH-IM lets you scan a QR code or enter a simple token to connect nine different platforms, like Telegram, Discord, and Slack, to a private AI harness on your computer. You can talk, share images, and manage tasks right inside your favorite chat window without needing any complex server setup or public IP addresses. It bridges the gap between casual messaging and serious coding work, keeping your data local while giving you instant access to powerful tools. It is the cleanest way to bring smart automation into your daily conversations.

🆔 @hackernewsgithubprojects

📦 nari-labs/nari-qwen3-tts

Nari Qwen3-TTS: Instant Speech Generation

Turn text into human-sounding speech in under fifty milliseconds, all running on just one high-end graphics card. Nari Qwen3-TTS is a specialized tool built specifically to run the Qwen3 speech model at incredible speed. The problem it solves is simple: most text-to-speech systems feel sluggish, forcing you to wait seconds for audio to appear, which makes building real-time voice apps frustrating. This project crushes that delay, delivering sound almost instantly even when handling many requests at once. It’s incredibly useful if you need fast, responsive voice features without managing a massive server cluster, proving that heavy AI models can be made to fly.

🆔 @hackernewsgithubprojects

📦 musoyangrigor/gitx-skill

GitX: The AI Skill That Cleans Up Your Messy Commits

GitX is the portable AI skill that turns chaotic agent edits into clean, logical Git history. If you have ever stared at a wall of code changes from an AI tool and panicked about how to commit them properly, this is your fix. GitX plugs into agents like Claude or Cursor to automatically group your work, write smart conventional commit messages, and even handle branch creation and pull requests for you. It handles the boring, risky parts of Git so you do not have to think about merge conflicts or secret scanning.

📰 https://news.ycombinator.com/item?id=49401487

🆔 @hackernewsgithubprojects

📦 westpoint-io/mimik

Record Your Workflow as You Click with Mimik

Hit record, perform your browser task, and watch a polished step-by-step guide assemble itself in real-time. Mimik is a local browser extension that captures every click and keypress to instantly create annotated screenshots, eliminating the tedious work of manual documentation. The standout feature is its Guide Me mode, which allows you to replay your recorded steps directly on the live webpage, highlighting the next element and guiding the viewer through the process without needing to leave the site. Because everything runs locally, your sensitive data never leaves your device, making it a secure way to document internal tools or create tutorials.

📰 https://news.ycombinator.com/item?id=48098187

🆔 @hackernewsgithubprojects

📦 neeeophytee/agent-stylebooks

Agent Stylebooks: Why AI Needs Editorial Rules

Agent Stylebooks is the collection of writing guides that stops AI from guessing at tone. Instead of asking a model to just sound human, you install a specific editorial system, like the one from the Centers for Disease Control or the government of the United Kingdom. The project takes proven public style guides and turns them into simple files that tell your coding assistant exactly how to structure a sentence. It turns vague instructions into clear, repeatable rules. This is useful because it means the error message in your app can sound exactly how a real support team would write it.

🆔 @hackernewsgithubprojects

📦 zackb/tether

Tether: iPhone Messages on Linux

Tether is the Linux bridge that finally brings your iPhone text messages to your desktop without needing a Mac. If you have switched to a Wayland-based system like Fedora or Arch but still rely on an iPhone, this project handles the heavy lifting for you. Instead of just syncing the clipboard, it pairs your phone over Bluetooth so you can read and reply to iMessage and SMS directly from your computer. It also mirrors notifications from any app, letting your Linux setup feel as connected as it used to.

📰 https://news.ycombinator.com/item?id=49370499

🆔 @hackernewsgithubprojects

📦 pgrundev/pgbot

pgbot: Your Database's Instant Health Check

pgbot is the one-shot health checker that turns your noisy Postgres database into a clear, readable report without installing a single agent. It connects with strict read-only permissions, scans the system statistics, and gives you a graded score with specific findings like slow queries or missing indexes. The coolest part is that it remembers the last check, so it tells you exactly what changed since your previous run, like a query that just got twice as slow. It is one static binary that works everywhere, making it perfect for quick triage or letting an AI agent get structured data. No servers, no setup, just answers.

🆔 @hackernewsgithubprojects

📦 smontlouis/bible-strong-avatar-lab

Build Animated Avatars Right In Your Browser

Drop a character into a browser and watch it come alive with expressive animations, all without touching a backend. Bible Strong Avatar Lab is a studio where you build procedural 2D avatars using simple shapes, define their facial expressions, and string them together into loops or one-shot clips. The coolest part is that once you design your character, you export a single portable file that works directly in React or plain JavaScript, completely detached from the editor. It runs entirely locally, meaning no accounts, no servers, and full control over your creative assets.

🆔 @hackernewsgithubprojects

📦 sapphiretoucan/diskgenius-shift

Diskgenius Shift: A Windows Workspace

Diskgenius Shift bundles a full team workspace into a single Windows download that requires zero installation. The project offers a calendar for scheduling, a document editor with markdown support, and an analytics dashboard to track team progress, all wrapped in a lightweight package that runs directly from your disk without needing extra dependencies. It is designed for simple, resource-light daily use, making it an interesting alternative for those who prefer a self-contained tool over complex cloud suites. The main takeaway is that this specific build provides a complete, offline-ready productivity stack for Windows users who want straightforward control over their workflow without the overhead of traditional software installs.

🆔 @hackernewsgithubprojects

📦 samukashvili/lumen-ps

Lumen PS: Turning a Flatbed Scanner into a 3D Texture Maker

Lumen PS is the texture-scanning software that turns ordinary flatbed scanner captures into professional 3D material maps. By rotating a subject like a leaf or circuit board four times, it uses the scanner's built-in light and sensor offset to calculate surface depth and color without any special camera rigs. The most surprising part is that it recovers detailed relief, such as individual veins or drilled holes, from just four simple scans. This solves the problem of needing expensive equipment to create realistic, relightable textures for digital art. It gives you normal, height, and color maps ready for game engines.

📰 https://news.ycombinator.com/item?id=49247297

🆔 @hackernewsgithubprojects

📦 cristianolivera1/aura

Aura: The Gradient Tool That Makes Backgrounds Feel Like Light

Aura is a web app that turns simple color layers into backgrounds that actually feel like glowing light. It ships over two hundred preset gradients built with CSS blend modes, so instead of flat fills you get soft, atmospheric washes that look like light bleeding through glass. The standout feature is the in-browser generator: stack layers, tweak blur, opacity, and blend modes in real time, add a grain overlay for a filmic finish, and even generate a prompt to rebuild the exact gradient in AI tools.

🆔 @hackernewsgithubprojects

📦 tgo-app-dev/vpipe

Vpipe Runs Massive AI Video Models on a 16GB MacBook

Generate video with sound locally on a standard sixteen-gigabyte MacBook Air. This is the surprising part. Vpipe is a local runtime built specifically for Apple Silicon that runs massive AI models, like the thirty-three billion parameter MiniMax H3, without needing a high-end graphics card or cloud server. It achieves this through custom Metal inference kernels and clever weight streaming, allowing you to create videos and their matching soundtracks entirely on your laptop. The entire runtime is incredibly compact, weighing in under thirty megabytes, which means it fits easily on your machine while handling heavy multimodal tasks like image editing, real-time chat, and video generation.

📰 https://news.ycombinator.com/item?id=49397410

🆔 @hackernewsgithubprojects

📦 cayu-dev/cayu

Cayu: The Runtime for Real AI Agents

Cayu is the production-grade runtime that actually lets AI agents survive long, complex jobs without falling apart. Most agent tools handle the quick chat fine, but they crumble when your agent runs for hours, loses connection, or needs to pause for a human approval. Cayu fixes this by making every step of your agent’s work durable and replayable. It treats your agent’s memory and actions like a real transaction, so if a server crashes, the agent picks up exactly where it left off. It also strictly controls what tools your agent can use and how it spends money, giving you a clear audit trail of every move.

🆔 @hackernewsgithubprojects

📦 liguobao/deepseek-harness-remote

DeepSeek Harness Remote: Control Your AI From Anywhere

Your work computer keeps running your AI coding agent, even when you are on a train, and this tool lets you check in from your phone. DeepSeek Harness Remote creates a secure bridge so you can follow active sessions, send new instructions with images, and answer permission prompts from your Android phone or another browser without moving your files. It works by turning your desktop into a host that connects outbound over an encrypted channel, meaning no public ports are opened and your data stays private. The project solves the problem of being stuck away from your workstation while your agent is still processing complex tasks.

🆔 @hackernewsgithubprojects

📦 r44vc0rp/pgrok

pgrok: Own Your Tunnel

pgrok is the self-hosted tunneling tool that gives you instant HTTPS on your own VPS. Forget renting from a third party to share your local dev server with the world. With one command, it spins up a secure SSH tunnel and Caddy handles the SSL certificates automatically, so any subdomain you pick works instantly. You even get a live terminal dashboard to watch requests flow through in real time. It is the perfect way to test webhooks or share a staging environment without exposing your home network or paying for a service you do not control.

🆔 @hackernewsgithubprojects

📦 only-cli/oc

Stop Wasting Tokens on Web Browsing

Turn any website into a compact command line interface for AI agents, cutting browsing costs from tens of thousands of tokens to just hundreds. This tool fetches a page and returns a tidy, numbered view, allowing agents to read, follow links, and search without the massive overhead of raw HTML. It even slips past basic blocks by mimicking a real browser. The single most impressive thing is that it lets an agent browse a live site using less data than a single tweet, making complex web navigation incredibly cheap. If you are building AI tools, this is the fix for your token bill. Check it out.

📰 https://news.ycombinator.com/item?id=49367419

🆔 @hackernewsgithubprojects

📦 raiyanyahya/llmaker

llmaker: The One-Command LLM Stack

A single terminal command can spin up a private, production-ready AI assistant, complete with vector databases and caching, without writing a single line of configuration glue. This project, called llmaker, solves the painful problem of manually wiring together different AI tools by provisioning the entire fleet on a private network from one simple file. It handles the networking, service discovery, and GPU management so you can focus on building applications instead of managing containers. Because everything runs locally, your sensitive data stays on your own hardware, and you can talk to the model using standard interfaces.

📰 https://news.ycombinator.com/item?id=48726917

🆔 @hackernewsgithubprojects

📦 techomancer/iris

IRIS: Emulating a 1990s SGI Workstation

Boot a full 1990s SGI Indy workstation on your modern computer using Iris, a new emulator written in Rust. It does not just mimic the screen; it simulates the actual MIPS processor, letting you run the original IRIX operating system with a working desktop, shell, and network access. The project is particularly interesting because it uses a specialized compiler to translate the old machine code into native instructions on the fly, ensuring the system runs fast enough for daily use. It even includes safety features that protect your hard drive images from crashes. This is a fascinating look at how complex legacy hardware can be revived on today's systems.

📰 https://news.ycombinator.com/item?id=47940203

🆔 @hackernewsgithubprojects

📦 hars-21/reqsh

The Persistent HTTP Shell

Reqsh remembers your API session even after you close the terminal. Built in Rust, this interactive shell saves your base URL, headers, and variables to a local file, so you pick up exactly where you left off without retyping anything. It acts like a smart command line for testing web services, letting you build requests line by line and instantly replay them. Because it handles state persistence automatically, it solves the annoyance of losing your test setup between restarts. It is a fast, lightweight tool for developers who need a quick way to hit APIs without writing code.

🆔 @hackernewsgithubprojects