Achieved AGI internally. #AGI

London ! We’ll be at Dust's Engineering Night on Feb 23. Vic (Vic Klein) from our team will speak. Agents will run. Register here: luma.com/w5keu427

The optimal recursion depth isn't clear to us yet, but we are glad to have folks like Omar Khattab driving research forward. Excited for what's to come.

Reasoning. Persistent state. Recursion. In our ARC-AGI implementation, agents autonomously decide when and what state to pass into a sub-agent’s REPL, allowing them to focus on analysing training examples, test inputs or both. Check out the logs github.com/symbolica-ai/a….

Funnily enough I tried to dabble with ARC AGI before and with very little success… Super cool to well designed RLMs achieving SOTA :)

Vic (Vic Klein) will chat ARC-AGI-2 results at Engineering Night London - what worked, what didn’t, and what surprised us. With Agentica: • 85.28% with Opus 4.6 (120k) High • +10pp GPT-5.2 (XHigh) vs CoT • +20pp Opus 4.5 vs CoT See you there. Register:

👀

Really cool to see teams already working on ARC-AGI-3 We'll have public replay links for every online run

I tried the ARC-AGI-3 test set, it’s quite well done, but I am sure AI models will solve it before year's end. It looks like it began before it was released ☺️. I suggest they already plan to release ARC-AGI-4 soon!

We have now solved all publicly available ARC-AGI-3 puzzles.🧩

Join us this Thursday, 26th in London. We’ll be speaking about ARC-AGI-2 and ARC-AGI-3 at the London AI nerd meetup: what’s new, what’s changing, and what it means for agents. Vic Klein Fireworks AI luma.com/z4i401h6