I love this opensource project! Someone just put a... 28.9M parameter AI model on an ESP32-S3. A chip that costs only a few dollars. No cloud. No API. No internet. It runs locally at ~9.5 tokens/s and can generate stories on a tiny screen. This changes the way we think about AI hardware. The next generation of AI toys, robots, and IoT devices may not need the cloud at all. Small chips are getting smarter. 🤖show more

for everyone who still doesn’t believe you can run... local AI on an ESP32: here’s the same class of chip booting an entire operating system. Windows XP boot screen. welcome screen. the Bliss wallpaper. and, of course, an error dialog😁 we went from fitting a 28.9M-parameter language model into an ESP32-S3... to recreating the full Windows XP experience on a tiny 240×240 display. the era of "microcontrollers are only good for blinking LEDs" is overshow more

forget the $699 AI pins. this $8 chip just... shattered the barrier for local AI hardware. a developer just forced a 28.9 million-parameter LLM onto a standard ESP32-S3 microcontroller. it costs roughly 8 dollars, runs completely offline, and draws the power of a single LED. conventional wisdom said a model of this size simply would not fit. the chip only has 512 KB of fast SRAM and 16 MB of flash. the breakthrough is architectural. the developer moved the bulk of the embedding table into flash memory and memory-mapped it. the chip only needs to pull about 450 bytes per token, keeping the active working memory inside the fast SRAM. this means you can now embed a capable language model into a physical node for the price of two coffees. and we are already seeing the beginnings of this custom physical hardware. in the video, a creator built a minimalist voice-controlled universal remote using an ESP32. it captures voice and remotely controls the computer over bluetooth LE. he simply says "open chrome and open 20 new tabs", and the custom hardware executes it instantly. we have spent years watching model sizes explode upward. but the true frontier is the opposite direction. when an eight-dollar chip can power offline intelligence and custom physical interfaces, AI becomes local infrastructure rather than a cloud service.show more

🚨 AN UNCENSORED AI MODEL THAT FITS ON YOUR... PHONE💀💀 Meet LFM3-1.7B Uncensored a lightweight local AI model designed to run on everyday hardware. What makes it interesting: Around 1 GB for a quantized version Up to 256K context supported by the model family Runs locally without a cloud API No monthly subscription or per-token fees Community-tuned for fewer content restrictions You can experiment with it using compatible local AI apps even on a phone with enough available memory. Note: The 1 GB figure depends on quantization, and actual performance varies by device. This specific model is text-only; vision and audio require separate models. 🔗 Model:show more

Holy shit... Microsoft open sourced an inference framework that... runs a 100B parameter LLM on a single CPU. It's called BitNet. And it does what was supposed to be impossible. No GPU. No cloud. No $10K hardware setup. Just your laptop running a 100-billion parameter model at human reading speed. Here's how it works: Every other LLM stores weights in 32-bit or 16-bit floats. BitNet uses 1.58 bits. Weights are ternary just -1, 0, or +1. That's it. No floats. No expensive matrix math. Pure integer operations your CPU was already built for. The result: - 100B model runs on a single CPU at 5-7 tokens/second - 2.37x to 6.17x faster than llama.cpp on x86 - 82% lower energy consumption on x86 CPUs - 1.37x to 5.07x speedup on ARM (your MacBook) - Memory drops by 16-32x vs full-precision models The wildest part: Accuracy barely moves. BitNet b1.58 2B4T their flagship model was trained on 4 trillion tokens and benchmarks competitively against full-precision models of the same size. The quantization isn't destroying quality. It's just removing the bloat. What this actually means: - Run AI completely offline. Your data never leaves your machine - Deploy LLMs on phones, IoT devices, edge hardware - No more cloud API bills for inference - AI in regions with no reliable internet The model supports ARM and x86. Works on your MacBook, your Linux box, your Windows machine. 27.4K GitHub stars. 2.2K forks. Built by Microsoft Research. 100% Open Source. MIT License.show more

Your PC can now run an entire AI studio... for free. Locally Uncensored puts chat, image generation, video generation, and a coding agent in one desktop app for Windows and Linux. - Chat with local models and your own documents - Generate images and videos on your GPU - Write code and review changes before applying them - Run agents with tools and MCP connections Install it, pick a model, and start. It shows which models fit your hardware before you download. No Docker. No terminal. No cloud required for the local workflows. 100% Free. Open Source.show more

Cancelled ChatGPT -> Built JARVIS -> Pays $0 ->... it works offline + it's smarter than the $20/month version. No WiFi needed, no cloud, no API keys, no rate limits, no queues, no $20/month just to ask a server in Virginia for the weather. Just a local model running directly on the laptop hardware, voice activated, system integrated, controlling apps, answering questions, doing the work. Iron Man had JARVIS embedded in his suit, this guy has it embedded in his MacBook and it works on a plane, in a basement, on a remote cabin with zero signal. OpenAI is burning $700,000 a day on infrastructure to deliver something this guy runs for free. Anthropic charges $200/month for unlimited Claude access, microsoft built Copilot into every product they sell. This guy skipped all of it, downloaded a model and made his laptop the smartest device in the room. No subscription. No login. No internet. No data sent anywhere ever. The most powerful AI assistant on earth is now the one running locally on hardware you already own. ChatGPT charges you to think slower, he pays nothing and thinks alone, he made it himself.show more

🤯 What, Chrome has an AI summarizer API built... in now?? - No API keys, no internet. Runs locally - Downloads the model on-demand - Available as a global in the latest version of Chrome. Works in chrome extensions too I have so many chrome extension ideas now...show more

I trained a tiny language model to write bad... tech jokes and run on a microcontroller 19M parameters 11 MB runs on a conference badge (RP2350), no internet, about 2 tokens a second it learned English from half a billion tokens of tiny kids stories, then I trained it on about 22,000 jokesshow more

Gemma 4 looks at a parking lot. Decides what... to ask. Calls SAM 3.1. "Segment all vehicles." 64 found. "Now just the white ones." 23 found. One model reasoning and orchestrating. One model executing. Both running locally on a MacBook. MLX. No cloud. No API.show more

THAT $70 "RUN YOUR OWN LLMS" PI KIT CAN'T... RUN A SINGLE LLM. IT'S A VISION CHIP WITH NO RAM. that clip sells a raspberry pi 5 in a slick case with an ai accelerator and the caption "your own llms." clean build, fun kit. the claim is where it breaks. the fine print: the popular $70 pi ai kit uses a hailo-8l, 13 tops. it's built for vision, object detection and image processing, and it has no memory of its own. so it cannot run large language models. full stop the board that actually can is a different one: the newer ai hat+ 2, hailo-10h, 40 tops, with 8gb of dedicated ram. that's $130, not $70 and even that runs only tiny models. llama 3.2 at 1b, qwen 2.5 at 1.5b, deepseek r1 at 1.5b. edge llms live in the 1-7b range, against cloud models at 500b to 2 trillion so the honest pitch: for $130 you can run a very small language model on a pi, slowly, as a fun learning project. that's real and it's cool. "your own llms" on a $70 vision kit is not. why this keeps happening: "ai kit" and a big "tops" number sell. tops sounds like intelligence. but tops measures vision-style math, not whether the chip has the memory to hold a language model. the spec that matters for llms is ram, and the cheap kit has none. the honest caveats, both ways: the $70 kit is genuinely great, just at vision. cameras, object detection, that's its job the $130 hat really does run small llms locally, which a pi couldn't do at all two years ago. that's progress "small" is the load-bearing word. don't expect gpt at home on a pi the takeaway: before you buy a kit because the caption says llm, check two numbers. not the tops. the ram, and the size of the model it can actually load. no 70-dollar miracle, no gpt in a pi case, no tops number that means what you think. save this before you buy the wrong kit for the word on the box.show more

A guy built a Pirate Bay for AI models.... - Open-source and leaked AI models. - Models for coding, search, and translation. - AI agents ready to download. - OCR models for photos and documents. - Runs locally, no cloud, no subscriptions. -show more

🚨GOOGLE CLOUD AND SUI ARE BUILDING AI PROOF! MystenLabs.sui... and Google Cloud are building Verifiable Agent Arbiter, a record that shows an AI agent stayed inside the limits a company set. The logs stay private in Google Cloud. The proof gets stored on Walrus and coordinated on Sui. Enterprise AI audit traffic landing on Sui.show more

AI VIDEOS ARE BECOMING SCARILY NORMAL At first, this... looks like a simple clip from a sunny day on campus. A girl is sitting on the grass with a notebook, people are relaxing around her, and the whole scene feels like something someone casually filmed on their phone. Nothing about it instantly feels fake. And that’s exactly the point. AI video is no longer only about creating robots, fantasy worlds or crazy impossible scenes. It’s getting better at copying the most normal parts of real life: sunlight, background movement, random people, small gestures and natural-looking camera shots. Not long ago, AI clips were easy to expose. The face would feel off, the hands would glitch, or the motion would look weird after a few seconds. Now those signs are becoming harder to catch. The real breakthrough isn’t that AI can make something shocking. It’s that it can make an ordinary everyday moment look so real that most people would scroll past without questioning it.show more

This is one generation. No retouch, no upscaler, no... cleanup pass. Look at the eyes, the catchlight, the lashes, the bit of moisture. That's the stuff that usually outs an ai shot, and at 4k it just sits there holding up. Enterprises and Developers can access the official API via BytePlus, and you can also use it directly on BytePlus Lumina. BytePlus Luminashow more

What If Your Phone Became the Screen for an... ESP32-S3? Droidputter lets you plug an ESP32-S3 into an Android phone over USB-OTG and use the phone as its display, keyboard, GPS and firmware flasher. The ESP32-S3 still runs the application. The phone simply provides the missing hardware. The clever part is that Cardputer apps can be rebuilt with a display and keyboard shim, while pixel data is streamed over the ESP32-S3's native USB-CDC connection. The phone can then flash the firmware and mirror the app. You can even take an open-source Cardputer app from GitHub and build a phone-mirrored version on demand. It is a really interesting combination of ESP32-S3, USB-OTG, Android, PlatformIO and open-source apps. Would you build an ESP32-S3 device that uses your phone as its screen?show more

AI can alr generate ads that no one can... tell you just need to use the right tool, workflow, prompts and a few bucks subsshow more

Gemma 4 watches raw video. Understands the scene. Then... prompts SAM 3 to segment and RF-DETR to track. One AI directing two others. Fighter jets. Crowds. Aerial defense footage. All three models running locally on a MacBook. No cloud. What scene should I point this at next?show more

THIS ENGINEER SKIPPED 7,862 WORDS OF AI ROADMAP AND... SHIPPED ON A WEEKEND a Pi Zero 2W, an IMX500 camera and a solar panel, that's the whole stack inference runs on-chip so there's no cloud, no latency and no monthly bill Spotted Dove at 0.95, King Parrot at 0.88, runs forever on sun code is open source and the 3D files are free, you just print the case and buy the parts and it works better than most paid wildlife monitoring services out there full video: code (open source): 3D print files: solar power board: save this so you actually have everything in one place when you decide to build itshow more

Run Uncensored Qwen-Image 2.1 on 16GB Mac Qwen's image... model with no filters makes the images cloud apps refuse and your prompts never leave your Mac, all of it done locally Run AI models locally ->show more