Skip to content
TopicTracker
From HackerNewsView original
TranslationTranslation

ZCode: GLM-5.2's own harness is officially live

ZCode, the official evaluation harness for the GLM-5.2 model, has been officially released and is now live for use.

Background

- ZAI (Zhipu AI) is a major Chinese AI company behind the GLM series of large language models, a leading alternative to GPT and Llama in China. - GLM-5.2 is ZAI's latest flagship model. "GLM-5.2's own harness" means the model now has its own native evaluation/testing framework, called ZCode, rather than relying on third-party benchmarks. - This matters because it signals GLM-5.2 is moving toward production readiness — a dedicated harness allows the developer team to run standardized tests for performance, safety, and capability tracking. - The context: Chinese AI labs (Zhipu, Baidu, Alibaba, DeepSeek, Moonshot) are racing to catch up with or differentiate from Western frontier models. Releasing a custom harness suggests ZAI is serious about systematic benchmarking and developer tooling around GLM-5.2.

Related stories

  • The author tests the new open-weights GLM 5.2 model against Gemini 3 Flash by having them play text adventure games within a fixed budget of $0.15 per attempt. A harness tracks achievements earned in each game to compare the models' performance.