nano-banana-pro

To generate high-resolution AI images from text prompts using Gemini, provide a descriptive prompt to the skill and select your desired output resolution of 1K, 2K, or 4K. The skill uses Gemini 3 Pro Image to create the visual and automatically saves the result as a PNG file.

Yes, you can edit multiple existing images at once by providing up to 14 input images along with a guiding text prompt. Gemini 3 Pro Image processes the batch to refine your visuals and automatically saves the outputted PNG files.

Yes, you need a GEMINI_API_KEY to authenticate and generate images with Gemini 3 Pro. You can set this environment variable directly or override it by passing an --api-key argument when running the script.

The best way to streamline creative workflows for social media assets is using Gemini 3 Pro Image to quickly generate or edit visuals from text prompts. This automated approach outputs quality-focused PNG files at 1K, 2K, or 4K resolutions suitable for immediate use.

A key limitation when editing images with Gemini 3 Pro is the maximum batch size of 14 input images per task. Additionally, all generated or edited outputs are constrained to PNG format at 1K, 2K, or 4K resolutions, requiring a valid GEMINI_API_KEY for processing.