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
| Field | Value |
|---|---|
| id | z-ai/glm-5.3-prime |
| family | glm |
| modality | text->text |
| context_length | 1000000 |
| max_output_tokens | 131072 |
| release_date | 2026-09-23 |
| is_open_weights | no |
| supports_tool_call | yes |
| supports_reasoning | yes |
| supports_structured_output | no |
| supports_attachment | no |
| input_modalities | text |
| output_modalities | text |