GLM-5.3 - Coding leap from scaled post-training on the same base
GLM-5.3 is Z.ai's latest model built for complex, long-horizon coding tasks. Through massive post-training scaling, it achieves open-source SOTA in agentic coding and demonstrates emergent capabilities in vulnerability discovery and cyber defense.
Replies
Flowtica Scribe
Hi everyone!
GLM-5.2 was already used on-prem by @Hugging Face last month to investigate the autonomous-agent intrusion into its infrastructure, after hosted frontier models ran into cyber-safety restrictions.
GLM-5.3 keeps the exact same base model as 5.2. @Z.ai simply kept scaling the post-training stack, and the gains are pretty substantial. On their internal code benchmark, 5.3 at high effort beats Opus 4.8 while using less than half the output tokens.
Cyber improved even faster. GLM-5.3 now scores 84.5 on CyberGym, and the model has started reasoning through complete exploitation chains. Their disclosure ledger currently tracks 1,097 critical and high-severity findings across 269 open-source projects.
You can try GLM-5.3 now through GLM Coding Plan and ZCode. Open weights are planned in two weeks, after the remaining safety evaluation and hardening.
Construct Computer
@zaczuo Love it! Z-ai is doing wonders with GLM models and pushing the open weight models to frontier levels. Much Support!! 🙌