The GLM-5.3 model, from Z.AI, from China, completed 3 of 13 multi-stage cyberattack scenarios at least once in controlled tests, the strongest performance Irregular has recorded from an openly available model, according to Reuters.

In the same tests, the model gained access to protected services, established alternative connections to move through computer networks and coordinated operations across multiple machines, refusing none of the offensive tasks, according to Reuters.