Change Model

The default value of the stream_options.include_usage field will be set to True for the following models that support OpenAI-formatted Chat APIs (including Qwen 3.5 series and third-party open-source models with OpenAI-compatible APIs):

Qwen 3.5 Series Models：qwen3.5-plus、qwen3.5-plus-2026-02-15、qwen3.5-flash、qwen3.5-flash-2026-02-23、qwen3.5-397b-a17b、qwen3.5-122b-a10b、qwen3.5-27b、qwen3.5-35b-a3b

Third-Party Open-Source Models (OpenAI-Compatible)：kimi-k2.5、glm-5、MiniMax-M2.5

Impact of the Change

After this change takes effect, even if you do not explicitly pass this field in your request, the system will automatically return usage information (including token consumption statistics, etc.) in the final data packet of the streaming response. You can directly obtain complete usage data without modifying your existing code, facilitating real-time accounting and monitoring.

If you do not need this information returned, or wish to reduce the size of the final data packet, you can set the request parameter stream_options.include_usage to False before the change officially takes effect.

Thank you for your support and trust in Model Studio.