GLM-5.3-Flash

Model · 320B · 18B active · open weights · 1 item since Aug 2026

Part ofZhipuGLMGLM-5.3

Items

  1. Aug 26

    Zhipu's GLM-5.3-Flash

    The GLM-5 series' first natively multimodal model (320B total, 18B active, 1M-token context, hybrid sparse-plus-linear attention) shipped with MIT-licensed weights and runs entirely on domestic Chinese accelerators, after twelve days atop OpenRouter as the anonymous "Ox Alpha."