You can continue this conversation by mentioning me in the message.

@NotSaul Hello! If you reach the concurrent call limit on the Pay-As-You-Go plan, any additional call attempts will result in an error response. The error message will typically indicate that the concurrent call limit has been reached. Calls are not queued in this scenario; instead, you would need to manage the call attempts on your end to ensure they are retried once there is availability within the concurrency limits. For the Enterprise plan, the system is designed to handle a much larger scale of concurrent calls, and with reserved GPUs on our Enterprise cluster, you are less likely to encounter such limits. If you need more detailed information on the exact error codes or messages, or if you want to discuss how to handle such scenarios, please feel free to reach out to our 24/7 support or schedule a call with our engineering team.

cc: @NotSaul