Feature Requests

Please adding support for Google's Gemini-3-Pro-Image-Preview (Nano Banana Pro) model, or at least parsing its key parameters: Aspect Ratio: e.g., "aspectRatio": "16:9" (supports 1:1, 16:9, etc.). Image Size: e.g., "imageSize": "4K" . Sample config: ```json { "imageConfig": { "aspectRatio": "16:9", "imageSize": "4K" } } ``` This would improve analytics for image workflows.

OpenAI has a model param called service_tier that accepts strings like "flex" or "priority". Currently your UI only allows numbers as valid inputs.

When using AI SDK from Vercel, when providing a zod schema when doing object generation it will always output the expected structure regardless if the model supports it or not. I assume they detect if the model natively supports structured output and uses that, otherwise just injects the schema into the prompt. Right now, if someone working on prompts has everything working well with structured outputs, but we has to use another model, they would have to copy and paste that into the prompt manually. Maybe a good way to do this is to define a default template for how the json schema will be injected and make that customizable as well.

The ability to include a knowledge base for testing to see what changes affect output. Also helpful for testing various prompts with one consistent knowledge base. Would also be useful to see what to include in the prompt to ensure responses are accurate based on the knowledge base.

The ability to include a knowledge base for testing to see what changes affect output. Also helpful for testing various prompts with one consistent knowledge base. Would also be useful to see what to include in the prompt to ensure responses are accurate based on the knowledge base.

Just Dark Mode for the UI

Lot of prompts get logged and it's useful to delete some old, unnecessary prompts from the sidebar

It would be great if there was a way to configure Gemini to use search grounding

please support AWS Bedrock Nova Micro Model. Quite crucial for us

It would be great if there was a way to create API keys with limited access. In my case, I would like my app servers and developers to pull in prompts from PromptLayer, but the API key used to do that should not be able to update the prompts (in case of compromise, etc).