April 2026: Elevated 5xx Errors & Timeouts for Instant Learning (LLM Nano)
Nanonets experienced elevated 5xx errors and timeouts for Instant Learning models using LLM Nano due to GPU service provider infrastructure issues. The team mitigated the problem by temporarily routing LLM Nano traffic to the more stable LLM Mini variant, which improved file processing speed and reliability. The incident was fully resolved after 51 minutes.