Zhipu AI Launches GLM-5.3 with 50% Improvement Over Predecessor

Zhipu AI has released GLM-5.3, claiming it to be the most powerful open-weights coding model, with a 50% performance improvement over its predecessor achieved through post-training. The model has identified 2,436 vulnerabilities across 269 projects, highlighting its utility in security assessments. The model weights are set to become open source in two weeks, allowing broader access for developers and researchers. This follows Zhipu AI’s ongoing efforts to enhance coding models in a competitive landscape that includes offerings from Meta and OpenAI. According to The Decoder, this release positions Zhipu AI as a significant player in the open-source coding model space.