I am running the Kokoro-82M model in a Google cloud function. The function gets executed between 5-10 times in a day, but I am still getting a 429 rate limit error:
huggingface_hub.errors.HfHubHTTPError: 429 Client Error: Too Many Requests for url: https://huggingface.co/hexgrad/Kokoro-82M/resolve/main/config.json
Getting rate limited for 10 requests daily seems unreasonable to me. I tried authenticating my session with a hugging face token hoping I would no longer get rate limited, but it didn’t help with the issue. Do folks have tips on how to proceed?
I am slightly suspicious this has to do with the fact that the model is executing in the GCP environment. When I execute the same code from my local machine, I can invoke the code dozens of times in a few minutes without running into any rate limit issues.