ai
Zhipu AI Releases Open-Weight GLM-5.2 Model That Trails U.S. Systems on General Tasks
China’s Zhipu AI released GLM-5.2 on June 28. Researchers said the model matches Mythos performance in some bug-finding and cybersecurity tests while trailing U.S. systems on broader tasks.
EngadgetAudio version
Tap play to generate a narrated version.
Developing·Limited corroboration so far. This page will refresh as more sources emerge.
Zhipu AI released its open-weight GLM-5.2 model on June 28. Researchers reported that the system matches Mythos performance in certain bug-finding and cybersecurity scenarios. GLM-5.2 trails models from Anthropic and OpenAI on general tasks, The Verge reported.
The release narrows the capability gap between Chinese and U.S. AI systems. The U.S. government has restricted China’s access to advanced models such as Anthropic’s Mythos and Fable and to the hardware required to train and run them.
The Trump administration has identified models capable of identifying vulnerabilities as national security concerns. OpenAI unveiled GPT-5.6 and restricted access to the system. GLM-5.2 can be downloaded and executed on readily available hardware by any user.
The Verge reported that the open-weight format provides flexibility for power users while allowing operation with limited oversight.
