Image model
Nano Banana 2 supports prompt-led generation and one-source-image edits. The configured workflow uses 1K output; review faces, text, and product details rather than assuming exact preservation.
Google · 67 Credits per successful result
portrait and general image generation plus reference-guided product edits at 1K
Supported operations: Text-to-image generation · Upload-and-edit
Canvas ratios: 1:1 · 2:3 · 3:2 · 3:4 · 4:3 · 16:9 · 9:16. Editing may preserve input framing where required by the selected workflow.
Negative prompt: not supported
Portrait and general image concepts
Reference-guided product scene changes
Single-image edits with source review
Run this model
Choose a supported operation and variant using the model links on this page.
Add a prompt or upload one source image in Studio, then review the selected model and Credits.
Inspect the result before publishing; retrying is a separate task.
No model-exact MorphPix output is published on this page yet. The configuration and use cases describe available controls, not benchmark results.
Nano Banana 2 supports prompt-led generation and one-source-image edits. The configured workflow uses 1K output; review faces, text, and product details rather than assuming exact preservation.
These configured variants do not support a separate negative-prompt field. Put desired visual direction in the main instruction and review the result.