Generative image generation

MINDTRICKS creates images in the cloud using Google's Gemini image model, then stores them for your account. Optional refinement in the browser is covered in Image post-processing tools.

Where to use it in the app

  • Image workspace—overview cards for the generator, GIF builder, and gallery.
  • Image Generator—text prompts, optional reference images, aspect ratios, and styles.
  • GIF Generator—animated frames produced through the same generation pipeline.
  • Prompt library—save and reuse snippets by type (style, character, action, scene).

How generation works

Requests go through the MINDTRICKS backend, which calls Google's GenAI API using the gemini-2.5-flash-image model with GenerateContent, including aspect ratio via image config—matching Google's documented flow for native image generation (see link below).

Depending on the feature, the API supports text-to-image, generation with multiple reference images, and masked edits. Each job runs as an asynchronous task: images are uploaded to cloud storage and recorded for your customer account so you can reopen them in the gallery.

Official API reference

Provider capabilities and request semantics are documented by Google: ai.google.dev/gemini-api/docs/image-generation.