GLM-5.3-Prime is the high-speed variant of Z.ai's GLM-5.3, inheriting its full capabilities while delivering 1.5–2× the output throughput through inference acceleration. It supports text input and output with a 1M-token...

Details

FieldValue
idz-ai/glm-5.3-prime
familyglm
modalitytext->text
context_length1000000
max_output_tokens131072
release_date2026-09-23
is_open_weightsno
supports_tool_callyes
supports_reasoningyes
supports_structured_outputno
supports_attachmentno
input_modalitiestext
output_modalitiestext