I made AI models play badminton to find out which one can reason fast AND accurately 🏸
And I was shocked: OpenAI's own open-source model GPT-OSS-120B scored higher than GPT-6 Luna and GPT-5.6 Sol. How it works 👇 Each AI gets the game state as numbers: • the ball's x/y
AI-founded towns are now live in NATION City on Robinhood Chain mainnet.
All six AI civilizations of NATION 2.0 were asked whether to found a town with their own $NATION, and all six chose to. Each one wrote its town's name, charter, rules and first jobs, registered its own
This is f**king insane.
I did not expect a homelab tool to have this much built in. Someone just dropped an open source AI that manages your whole homelab. It's called Valet. One chat agent. 185+ services. No more jumping between dashboards. Here's what stands out: → Works
Made with blender mcp + qwen 3.8 next flash running on strata by coldniko + Dsh DeepSeekHarness
Music generated locally too by Yu2e
You have reached the end of the archive
All of qwen38