Hi NVIDIA Team,
I’m requesting a rate limit increase for my account from the current default (40 RPM) to 200 RPM.
Account: dritan-hoxha@live.com
Use case:
I’m running a Hermes Agent (NousResearch) instance on a Ubuntu VPS, using NVIDIA NIM as the primary LLM backend with DeepSeek V4 Pro for an automated crypto trading bot. Kimi K2.6 is configured as a secondary/fallback model. The bot performs continuous market analysis, signal generation, and trade decision workflows — each cycle triggers multiple sequential LLM calls.
Problem:
With the current rate limit, I consistently hit HTTP 429 errors during active trading sessions, even with retry logic configured (api_max_retries: 3 with exponential backoff). The bot requires approximately 80–120 RPM to function reliably during peak market hours.
Infrastructure:
- OS: Ubuntu (VPS)
- Framework: Hermes Agent v0.13+
- Primary model: DeepSeek V4 Pro via NVIDIA NIM
- Fallback model: moonshotai/kimi-k2.6
- Deployment: Single instance, non-abusive usage pattern
Requested limit: 200 RPM
I’m happy to provide any additional details or verification needed. Thank you for your consideration.
Best regards