Z.AI Introduces GLM-5.1: An Open-Weight 754B Agentic Model That Achieves SOTA on SWE-Bench Pro and Sustains 8-Hour Autonomous Execution

Z.AI has released GLM-5.1, a 754B-parameter agentic model that achieves state-of-the-art performance on SWE-Bench Pro and demonstrates sustained autonomous execution for up to 8 hours. This matters to developers because GLM-5.1 can handle complex engineering tasks autonomously over extended periods, reducing the need for human intervention in iterative optimization processes.

Read the full article at MarkTechPost

---

Want to create content about this topic? Use Nemati AI tools to generate articles, social posts, and more.