-
Notifications
You must be signed in to change notification settings - Fork 2.2k
Description
I know there is support of disable/enable Gemini model thinking and we can control this in Gen AI Google SDK individually or with Spring AI support for Gen AI SDK as well.
However We have a system in which we are using Spring AI Vertex AI gemini with 1.0.0-M6 version of Spring AI and using Gemini 2.5 flash model.
I reasearch a lot but didn't found any option to disable the model thinking and it caused lot of latency in our workflow.
I am afraid it's available or not ? I researched and found it's not supported in Spring AI Vertex AI Gemini integration.
Is there any plan to introduce it ? If already there is something then please do let me know.
Expected Behavior
withThinkingBudget(0)
Current Behavior
No option to diable the thinking budget.
Context
It is impacting our workflow and causing lot of latency in our workflow.