GLM-5.2 Beats GPT-5.5: China's AI Excels in Coding

A new AI model from China, GLM-5.2, is outperforming OpenAI's GPT-5.5 on key benchmarks. This large language model, developed by Z.ai, is designed for long-horizon autonomous coding and engineering tasks. What's interesting is GLM-5.2 scored 54.7 on Humanity's Last Exam, beating GPT-5.5's score of 52.2. It also performed strongly in the FrontierSWE benchmark test, outperforming GPT-5.5 by 1%. The model features a stable 1-million-token context window, allowing it to handle complex engineering projects and complete full development workflows. Z.ai has released GLM-5.2 under an unrestricted MIT open-source license, making it freely available. They've also launched three coding plans for developers, ranging from $12.6 to $112 per month. This means more powerful, open-source AI tools are becoming available for developers.