Model
Nano Banana Pro
for AI agents
Nano Banana Pro is a better fit when an agent needs to edit an existing image instead of generating from scratch. Through AnyCap, the model becomes part of the same capability runtime and CLI used for other image and video tasks, so agents can move from generation to revision without changing tools.
Early Access
AnyCap is currently in early access. Capabilities shown on this page are available to early access users. Request access on GitHub to get started.
Best-fit workflow
- Revise an existing image with a text instruction
- Adjust lighting, composition, or background without rebuilding the asset
- Run fast agent iteration loops on visual drafts
- Start from an input image instead of a blank prompt
Example CLI call
Edit an existing image
anycap image generate --model nano_banana_pro --prompt "replace the background with a warm studio light" --image ./input.png
This is the kind of prompt an agent can issue after inspecting a draft, receiving feedback, or updating a design asset inside a larger workflow.
When not to choose it
Choose Nano Banana Pro
When the agent already has an image and needs targeted changes, multiple revisions, or prompt-based editing.
Choose Seedream 5 instead
When the workflow begins from a text prompt and the agent needs a polished fresh image rather than an edit pass.