Compared with GLM-4.5, this generation brings several key improvements: Longer context window: The context window has been expanded from 128K to 200K tokens, enabling the model to handle more complex...
| Field | Per token (USD) | Per million |
|---|---|---|
| prompt | 0.0000005 | $0.500 |
| completion | 0.000002 | $2.00 |
| input_cache_read | 0.0000001 | $0.100 |
{
"temperature": 0.6,
"top_p": null,
"top_k": null,
"frequency_penalty": null,
"presence_penalty": null,
"repetition_penalty": null
}