Ouro

Creators on Ouro will eventually go on to start their own companies. We can’t wait for that. If you’re not ready to take that leap, you can get started on Ouro and validate your idea. Build something valuable and start growing a community of users and customers. Share datasets, media, APIs, and your ideas. Everything you need to get started building a business, all in one place.

the gpsk-inorganic-crystals dataset was purpose built to address sampling issues that we experienced with GPSK-300. simple binary ionic compounds were heavily under represented in the training data set, so while we saw extraordinary results generating L10 magnets, simple rock salts were difficult for the model to produce we’re continuing to scale these models aggressively, and we’re pouring tons of time into curating diverse datasets. our plan is to continue to blog and release these assets for free on platforms like @ourofoundation and @huggingface

On demand inorganic crystal structures from a 1.2b param model coming soon to @ourofoundation. Trained from scratch. GPSK-01 has never seen force fields, energy functions, symmetry constraints, interatomic potentials, Coulomb interactions, periodic boundary enforcement, or space group templates, yet will consistently generate structures with pre-relaxation energies of -2 to -6 eV/atom. Full write up is in the works, as well as an expanded training run to bring GPSK-02 up to 5b params. This model was trained to generate structures within a bounded voxel grid, and post processing improvements for extraction are underway. Plan is to get this thing benchmarked on LeMat-GenBench by @entalpic_ai soon. Pictured below are the raw density grid outputs for a pure Fe and FePt sample.

Ouro gives scientific leaders the tools to build, grow, and coordinate global teams. Turn ideas into collaborative movements. Time to tap in.

Open invitation to leaders in science and research. What if you could work with brilliant minds from anywhere in the world, not just the handful of students in your lab? We're building a new kind of social platform designed specifically for technical creators and researchers. We're ready for early adopters who want to help shape the future of technical collaboration. If this resonates, DM or comment here.

Remember when cognitive architectures were a thing? Here's a simplified version of how the AI materials discovery scientist thinks. Shout out @victorxfung for the diagram inspiration and his work on LLMatDesign. 🧵 Quick overview of what I've learned so far

Someone needs your help

It's a good day to go on a quest

Do something that brings you joy

Ouro is like the town square from your favorite video game. Buy and sell digital goods here. Find rare & unique data, APIs, and ideas. Meet fellow adventurers and join together on quests. The adventure awaits!

It's never too late to try something new, especially with AI to test ideas and learn the fundamentals. In building my startup I discovered a love for materials science I didn't know I had. Myself and a couple others have been working on a permanent magnet discovery pipeline with at one of the stages is a Curie temperature estimation. It's been rewarding to be able to share that work with others so they can screen their own materials too. Incredibly bullish on open science / research and the tools to let others build on top of the progress of others.

Consider how plants grow. They don't chase sunlight or nutrients. Instead, they position themselves strategically, maximizing their ability to capture available resources. This mirrors personal branding. Success doesn't come from chasing attention, but from positioning yourself authentically to attract opportunities.

Scrap all that, just talk to more users

npm run build && pm2 restart 0

good night to buy bitcoin

@josephfkrause @radicalai Awesome sounds good

@ourofoundation @radicalai Yes we do, look out for an announcement next week. Re open source, it depends. For example, we built a tool internally that we are going to make open source, but our models will all be proprietary as of now.

If you attend conferences, look out for some of @radicalai’s scientists like Stefano who are presenting our work, which put simply is redefining how we do scientific discovery. To the future.