gemini-imagegen

Text-to-image generation with Gemini API converts text prompts into images by configuring the gemini-3-pro-image-preview model with your GEMINI_API_KEY, setting aspect_ratio and image_size in image_config, and submitting prompts through the API. The Skill automates this workflow and handles image asset output directly.

Yes. Image editing and multi-turn refinement let you modify existing images and iterate on results through sequential prompts. The Skill supports uploading reference images and maintaining context across turns to progressively refine outputs toward your target.

Gemini image generation supports configurable aspect_ratio and image_size settings through image_config. The Skill returns IMAGE response_modalities alongside TEXT output, and handles multimodal responses natively with built-in asset handling for generated images.

The Skill supports up to 14 reference images across iterative prompts for multi-image compositions. This enables complex collages, mockups, and product layouts where you provide multiple source assets to guide generation.

Gemini image generation requires a valid GEMINI_API_KEY and depends on the google-genai library and Pillow for image handling. Multi-turn refinement workflows are constrained by API rate limits and model capabilities; complex compositions with many reference images may require careful prompt structuring.

Yes. The Skill is purpose-built for logo, sticker creation, and product mockups through text prompts and image editing. Configurable aspect ratios and the ability to reference multiple source images support these specialized design use cases effectively.