Hi everyone,
I’m using Gemini API streaming in a chat UI, but the chunks I get are sometimes pretty large.
Is there any way to make the chunks smaller, or is this controlled by Gemini on the server side?
It would be nice if the API could stream smaller pieces of text for smoother UI updates.
Thanks!
Hi @Leo_Smith
Thank you for the feedback! Currently, the streaming chunk size is managed dynamically on the server side by our model serving infrastructure. This batching is designed to optimize network efficiency and overall throughput
I have forward this request internally as well
Got it, thanks for the clarification