Change Details

Effective 00:00:00 (UTC+8) on July 15, 2026, Model Studio will reduce the unit price for the Fast mode of the GLM-5.2 model. The details are as follows:

Fast Mode Overview

Fast Mode delivers higher Tokens Per Second (TPS) throughput, making it ideal for latency-sensitive scenarios such as AI streaming coding assistants, multi-step Agent reasoning, and real-time conversations. Please note that this feature is currently in Preview stage, and its capabilities and specifications may be subject to change with future versions. For more details, please refer to the Fast Mode Documentation.

Key Features:

• High-Speed Output: TPS is increased by 1.5x–2x compared to the standard API, reaching 80–100 TPS.
• Token-Based Billing: Billing logic aligns with the standard API, charging based on input and output tokens. For specific pricing, please refer to the Model Pricing Page.
• Flexible Throttling: Requests exceeding the TPM (Tokens Per Minute) quota are not immediately rejected but are instead placed in a queue for processing.

How to Enable:

Simply specify the supported Model ID in the model parameter when making API calls. No additional parameters or code changes are required.

Thank you for your continued trust and support in Model Studio.