Did Codex Reset
GitHub

Agora-2 lets up to 20 humans and agents share one real-time simulated world, elvis says

elvis

elvis says Agora-2, described as a new world model, now lets up to 20 humans and agents share one simulated world in real time.

elvis says the shared world can be used to train robots, self-driving cars, and cyber defense agents together, with each improving as the others get better. elvis also describes it as a safe place to study agent collusion before it reaches real systems.

#agora-2#world-models#multi-agent#simulation

Best Taste-Bench model chooses the better direction 59.7% of the time

A paper from Microsoft and colleagues reports that the best model answers 59.7% of Taste-Bench questions correctly when choosing the better open direction in long engineering and research tasks. Forks decided by later evidence are much harder, a larger reasoning budget does not raise accuracy, and distilling a teacher’s outcome judgment improves held-out SWE-bench Pro success.