benchmark
Benchmark for exploiting discovered vulnerabilities; GLM-5.3 reported 54.4 percent, trailing rivals.
Greg Brockman warns an open-weight release due at the end of August will worsen the threat landscape, while OpenAI's strongest cyber model sits behind identity checks and hardware keys.
Reality
Zhipu says cyber capability outran expectations during post-training, so downloadable weights slip to around August 28. Capability gating is now a management call, not a rule.
Perspective Coverage
3 publishersReality
Z.ai says GLM-5.3 edges Anthropic's restricted Mythos 5 at vulnerability discovery while losing badly at exploitation. On vendor numbers, the defensive half is commoditising first.
Reality
Zhipu says GLM-5.3 edged Anthropic and OpenAI on one security benchmark. On the harder exploitation test the gap runs the other way, by 23.6 points.
Reality
Z.ai says all of GLM-5.3's coding gains came from post-training on tenfold more long-horizon task environments. The uneven benchmark jumps tell you where that money actually landed.
Reality