Hello NVIDIA Support Team,
I would like to request an increase to the API rate limit for my NVIDIA NIM API account.
Current rate limit: 40 RPM
Requested rate limit: 200 RPM
Use case:
I am developing an autonomous AI agent application that uses NVIDIA NIM as its LLM backend. The agent performs multiple tasks through iterative reasoning, tool usage, context analysis, and validation.
Because of the agent architecture, a single user instruction can result in multiple API requests being generated within a short period of time. Depending on the task, the agent may generate approximately 20–50 requests in a short period, sometimes concurrently.
With the current 40 RPM limit, these bursts frequently result in 429 Too Many Requests errors, interrupting the agent’s execution.
Increasing the limit to 200 RPM would allow the application to operate reliably during these short bursts while remaining within reasonable development and testing usage.
This is currently for development and testing purposes.
I would appreciate it if you could review my account and let me know whether an increased rate limit can be granted, or what requirements I would need to meet to obtain it.
Thank you.