Errthang is alive

I developed an LTX 2.3 program based on the desktop version of LTX, with optimizations that bypass the 32GB VRAM limitation. It integrates features such as start/end frames, text-to-video, image-to-video, lip-sync, and video enhancement. The links are in the comments.

Lets go! Play pretext breaker 🎮 pretext-breaker.netlify.app

Just released heerich.js, a tiny voxel engine that renders 3D scenes to SVG 🎨 ╬ Boolean ops ◮ Oblique & perspective 𝑥 Zero dependencies ◌ Pure vector output, infinite scaling Named after Erwin Heerich's geometric cardboard sculptures.

Black Mirror S8E1: In 2027, developers are allocated a daily Claude token allowance by the government. A junior dev burns through his entire month's supply trying to centre a div. His family starve. He is forced to write the code himself. He can't. Society collapses.

llms can FIGHT now. here's opus as wizard vs gpt-5.4 as robot. calling this budok-ai. it works by modding the brilliant game yomi hustle. 8-model seeded tournament incoming. details and code below:

Claude got busy simulating the SNES PPU at the hardware level then corrupting it just to give lil' old me some nostalgic SNES glitches for my CRT display.

@RoyalCities I'm using it in my journey to making beats, learning a lot from the terminology and now I listen to a beat and think about the prompts 😂😆 Claude know how to prompt foundation1 well but still lost in translation when I want a Yeezus type beats

The response to Foundation-1 has been overwhelming. Didn’t know how it would land - AI in music production is...complicated. Infinite samples was a bit of a wild card. I wanted to show there’s a middle path where it enhances creativity. Appreciate you all ❤️

Ropedia Xperience-10M is out on Hugging Face a large-scale egocentric multimodal dataset of human experience for embodied AI, robotics, world models, and spatial intelligence It contains 10 million experiences (interaction) and 10,000 hours of synchronized first-person recordings with six video streams, audio, stereo depth, camera pose, hand mocap, full-body mocap, IMU, and hierarchical language annotations dataset: huggingface.co/datasets/roped…

I've long preferred Claude Code over Codex or Gemini, because it seemed much more reliable, but couldn't explain why : now Bullshit Bench by @petergostev provides compelling numbers. It measures bullshit as "when given false premises disguised in jargon, will the model go with the flow (=bullshit) or push back (=truthful)" And Claude is leagues ahead ! Also, this objective of truthfulness is probably at odds with the Chatbot Arena emergent objective of "pleasant chat experience" ; but a model optimizing for the former will be more useful.

Y'all are quick. ComfyUI node for Foundation 1 (even has my random prompt build logic!) github.com/SanDiegoDude/s…

After months of work, today I’m releasing Foundation-1. A SOTA text-to-sample model built specifically for music production workflows. It may also be the most advanced AI sample generator currently available - open or closed. • ~7 GB VRAM • Entirely local • 100% free 😁

#CVPR submission ✅ LLM lost in translation - yolo segmentation/pose - #sonnet 4.6 json payload via #openrouter - per frame commentary - renderer constructs the visuals The thoughts are LLM realtime attempts to understand what it's looking at.. not just classifying it

Game changer for compositors! Rotate an image as if it’s a 3D object in the #Photoshop beta! Interested??

@StaniKulechov ex- thesis, ex- blockfi here I want to do design engineering, to bridge that design/dev gap where things always break, it where I excel. lets chitchat

Aave Labs is strengthening its Product and Enterprise Sales teams, and we’re hiring with open roles! Senior Product Designer Staff Design Engineer Senior Design Engineer Staff Smart Contract Engineer Product Marketing Manager Sales Engineer Director of Enterprise Sales and Partnerships Business Development Associate We’re a product-driven company focused on bringing DeFi to millions of users and institutions around the world by building the most compelling DeFi products. Apply at aave.com/careers

I forked AutoResearch and adapted it to run on macOS using Metal. Welcome! github.com/miolini/autore…

Dashboard for AI agent orchestration github.com/builderz-labs/…

Karpathy's llama2.c showed you could train a real transformer in pure C with no frameworks. A solo researcher (and Claude Code) just took that same model, Stories 110M, Llama2 architecture, trained on real text and ran it on Apple's M4 Neural Engine (ANE) for less than a watt. He reverse-engineered the undocumented private APIs, bypassed CoreML, and found Apple's abstraction layer was hiding 2-4x of the chip's real throughput. The ANE delivers 6.6 TFLOPS per watt, roughly 80x more efficient than an Nvidia A100. The real implication here is inference: there are hundreds of millions of Apple devices with one of the most efficient AI accelerators ever shipped in consumer hardware, and Apple's own software stack is the thing standing between developers and its actual performance. h/t @maderix

I will reveal a few surprises soon ;) Original shader created by @XorDev

Ffmpeg and good old SAM2

One of my biggest super-powers is using AI to crop, edit, stitch, overlay and improve videos and images. It is wild I can do that. Re-motion & ffmpeg are insanely useful to AIs, I will never do it myself again, even local models do this well.