xAI API gRPC RESOURCE_EXHAUSTED: what it means and how to fix it

What it means

gRPC status raised by the xAI Python SDK when a resource quota has been exceeded, meaning the user has exceeded their API usage quota or rate limits. Sync clients raise grpc.RpcError and async clients raise grpc.aio.AioRpcError.

Causes

• The user has exceeded their API usage quota or rate limits for requests

How to fix it

1. 1
   Fix 1.

   Retry with exponential backoff on RESOURCE_EXHAUSTED (official xAI SDK example).

   Snippet

   Copy

   def request_with_backoff(prompt, max_retries=5):
       chat = client.chat.create(model="grok-4.7")
       chat.append(user(prompt))
       for attempt in range(max_retries):
           try:
               return chat.sample()
           except grpc.RpcError as e:
               if e.code() != grpc.StatusCode.RESOURCE_EXHAUSTED:
                   raise
               wait = 2 ** attempt
               time.sleep(wait)
       raise RuntimeError("Max retries exceeded")

2. 2
   Fix 2.

   The SDK does not retry this code by default; add it to a custom gRPC retry policy via the grpc.service_config channel option.

   JSON

   Copy

   "retryableStatusCodes": ["UNAVAILABLE", "RESOURCE_EXHAUSTED"] # Additional status code for retry