Entropy - Craft desktop apps with TypeScript (and no WebView) Yumon - Mixture of Experts model, iGPU-Ready, Trained on You https://teapawt.org/t.co/wHV62AENf1
Grand Rapids, MI
Joined March 2018
- Tweets9K
- Following573
- Followers583
- Likes3.9K
Pinned Tweet
Yumon is the future. Do you not believe that robots will someday live amongst us as regular civilians?
🤖 Made with AI
Yumon is already available in v0.1 on Indie Machine. It's slowly finding its way into the world. Right now there's literally a fish tank simulation. But I think Yumon has a lot more potential than that. Maybe if I can squeeze in early with this form of agent, I can carve a brand
Working with Chat to envision my next Yumon architecture. Each component will be an independent model, running in parallel, lightweight and focused, managed by a small Central Nervous Model.
compression has more power in ML than we may ever even learn. although AI itself will likely uncover it and implement it. a little knowledge on how video encoding or zip or JPEG compression works will help. easy wins include needless duplicate 3D data in each frame
imagine a loop where memories and real-time visuals are packaged into temporal frames, combining chain-of-thought and image recognition. Significant frames are evicted slowly. The point is real-time interaction in 3D space, so that's all 4 dimensions potentially in one RoPE
I regret not creating the agentic loop (code, compile, fix) before Code came around. But I didn't have distribution anyway
I'm looking more closely at my images. Comparing my paint-style approach to my photographic approach. Up-close, it's clear just how beautiful and coherent they have become. I'm so happy about these outcomes.
I'm having a lot of fun with Yumon, but I'm probably going to delay InstaCode. InstaCode with Kilo shows effectiveness at learning syntax, but ultimately, a utility tool contrasts with a Pet, as it requires broad knowledge and long-horizon reasoning.
Alex Goodman retweeted
📦 NVIDIA did it first, now we are releasing 2,600+ free game-ready 3D models (+200 each week). For good.
🎛️ Each model is a program, with parameters you can tweak to customize look, state, season, color, wear, etc.
All thanks to the $500,000 in funding we received from @xprize and @GeminiApp and all of our early individual funders, thank you all!
Go build.
polyfork.dev
NVIDIA just put 2,575 simulation-ready 3D assets on Hugging Face 🤯
A 472GB collection covering everyday objects, tools and industrial equipment.
It’s called NVIDIA SimReady Assets. The models include physics settings and semantic labels for use in simulated environments.
- Train robots to pick up and move objects
- Build virtual factories and warehouses
- Generate labelled training images
- Populate 3D scenes with ready-made objects
Useful for robotics research, digital twins and anyone building detailed virtual environments.
Download the collection below 👇
here's what is amazing: you can build a multi-modal system in which there is image recognition, image generation, text generation, memory management and even temporal data like video and audio, all on a lightweight iGPU. you can literally build an entire person in your garage.
not yet. no. but soon. the tiger will climb up and out from within his pixel lair, he will be great among the cats of the wild, for he knows of a world made only of color.
[epoch 11/100]
Alex Goodman retweeted
Our robots built this in the city of Almere: curved walls, different types of bricks and bonds, indented text.
Real world deployment, insane complexity, 100% of every brick done by our robots, with another 15 houses to come in the coming weeks.
Alex Goodman retweeted
No Man's Sky releasing today on iOS!!! ❤️
apps.apple.com/app/id1615755…
fighters feel tough when they go to war, but the greatest challenge of all time is getting an animal to lay down his weapons.
Alex Goodman retweeted
Today, in collaboration with Anthropic, we’re excited to share rapid initial progress towards an antibody treatment for food allergy, which affects over 200 million people worldwide, discovered with the help of Claude.
At Divergence Labs, we believe that AI can unlock the next generation of antibody therapeutics with more effective mechanisms than before. Such mechanisms are urgently needed in food allergy, which involve overcoming one of the tightest affinity protein pairs in immunology, IgE and its receptor FcεRI, on mast cells.
Treatment with Xolair—the only approved antibody for food allergy—sequesters circulating IgE, but mast cells remain sensitized for weeks to months due to the extraordinarily slow dissociation of IgE from FcεRI.
We envision a treatment that actively pulls IgE off of FcεRI, without waiting for natural dissociation. Like picking a lock, this requires the right combination of geometry and force.
Our research team, working with Claude, ideated on mechanisms to accomplish this disruption, and converged on an antibody format with two arms with different functions and epitopes. Within one month, we designed, synthesized, and experimentally confirmed the biophysical function of the two arms, which we are now assembling into full therapeutic formats.
the latest art is up top, with art from several days ago on the bottom row. the art from several days ago was strongly cherry picked, while the art from today rarely requires a second try
Yumon Paint genuinely looks like paintings which I think is the cherry on top. i do wish for higher fidelity, but maybe its not the right time
