Facing some serious lag in responses with Gemini 2.0 in audio modality in multimodal live API

Since last 4 days I have been facing some serious lags in the audio response from Gemini 2.0 in multimodal live API through python code using google-genai.
First I thought this is an issue with my code or the sdk but I checked it on Google AI Studio live stream and faced the same issue (https://aistudio.google.com/live.)

Anyone has an insight into this issue?

1 Like

Yes unfortunately I had the exact same experience – thought it was my code but was easily able to reproduce it on AI Studio. I was disappointed because gemini-2.0-flash-exp had been very good and very affordable compared to other options.