I'm encountering sporadic performance/reliability issues with the 2.5 Pro API. While my prompt is within token limits and safety settings are adjusted, I find that I must execute consecutive, back-to-back API calls to ensure a successful response, which significantly increases my operating costs. Is this the limitation with API?
Prompt Works in Gemini UI, Returns Null via API: Seeking Urgent Fix. I’m stuck with this issue and have a release in a week, Any help Appreciated!
