I am relatively new to using Google’s suite of products for AI inference. One of the more recent features I have been trying out is their Multimodal Live API using gemini-2.0-flash-exp. I have read over, almost, every piece of documentation to get the live API to work. It seems that the consensus is to use the following (all done in python 3.11.9). I am trying to achieve simple text to text generation before moving to video/video frame to text generation in real time. However, when I try to run the code, I receive the following error: ConnectionClosedError: received 1011 (internal error) Request trace id: 2f02bfds67892de45, [ORIGINAL ERROR] generic::resource_exhausted: RESOURCE_EXHAUSTED: Maximum concurrent se; then sent 1011 (internal error) Request trace id: 2f02bfds67892de45, [ORIGINAL ERROR] generic::resource_exhausted: RESOURCE_EXHAUSTED: Maximum concurrent se. I have checked to make sure that I have the correct API key and that all of my environment variables are set. However, the most interesting aspect of all of this is that this error is occurring upon the first time that I try to establish a connection to the live api websocket. Please keep in mind that I am new to this, any tips would be greatly appreciated.

Compiled with problems:
×
ERROR in ./node_modules/gaxios/build/src/gaxios.js 68:16-32
Module not found: Error: Can’t resolve ‘https’ in ‘E:React Projectnewsappnode_modulesgaxiosbuildsrc’

ERROR WHEN RUN THE CODE BELOW
raise GoogleGenerativeAIError(f”Error embedding content: {e}”) from e
langchain_google_genai._common.GoogleGenerativeAIError: Error embedding content: 404 models/embeddings-001 is not found for API version v1beta, or is not supported for batchEmbedContents. Call ListModels to see the list of available models and their supported
methods.