GLM5.3
Input, lowest
$1.4
Output, lowest
$4.4
Providers
1
Providers & prices
| Provider | Input | Output | Context | Last sync | |
|---|---|---|---|---|---|
D
DigitalOcean
|
$1.4 | $4.4 | 1,048,576 |
About this model
Z.ai's flagship large-scale reasoning model, built for complex software engineering and long-horizon agent tasks. A 753B-parameter Mixture-of-Experts model (40B active) using the same base model as GLM-5.2, with all improvements driven by scaled post-training: a 50% coding performance gain on Z.ai Code Bench, Terminal-Bench 3.0 up from 4.6 to 28.3, DeepSWE v1.1 from 46.2 to 66.9, and Agents' Last Exam from 23.8 to 28.5. Reasoning is always enabled with three effort levels (low, high, max). Text input and output only, 1M-token context window, 128K maximum output.