Generate visuals from prompts
Describe the desired image using subjects, scenes, lighting, composition, and style. Start with 1K to compare directions, then choose 2K or 4K based on the actual display size.
Nano Banana 2.1 is suitable for creating images from text concepts, or progressively adjusting visuals with reference images and instructions. Choose 1K, 2K, or 4K resolution based on delivery needs, and complete both generation and editing in the same API.
Submit requests to the public API at api.acedata.cloud using the documented parameters, then use the results in your application.
api.acedata.cloudnano-banana-2.1Input parameters and result formats vary by service. Use the public API for this model and follow its guide for generation, task retrieval and editing operations.
Specify the full model=nano-banana-2.1; without the :official suffix. Request and response fields vary by endpoint; refer to the respective API documentation.
Describe the desired image using subjects, scenes, lighting, composition, and style. Start with 1K to compare directions, then choose 2K or 4K based on the actual display size.
Provide accessible image links, and clearly specify what to change and what to preserve. Still inspect key objects, text, and brand details in the edited result.
Choose square, landscape, or portrait proportions; resolution determines output detail, while aspect ratio determines composition, and they can be specified independently.
First generate or upload a subject image, then specify a desktop, indoor, or outdoor background to compare multiple presentation directions.
Choose a landscape or portrait aspect ratio based on the placement, and reserve space for titles and interface elements.
Use image editing to adjust colors, local objects, or visual style while preserving structures from the original image that need to continue.
2.1 is the successor model and can serve as a new efficient starting point for creation. Both support high resolution; existing fixed workflows can validate results separately using the same prompts.
Pro is designed for more complex finished visual work. For strict brand consistency, multi-element layouts, or critical text, it is recommended to try both separately and review the results manually.
For creating from scratch, use action=generate; to modify an existing image, use action=edit and provide image_urls.
Provide model=nano-banana-2.1 and prompt, and set aspect_ratio and resolution=1K/2K/4K as needed.
For synchronous requests, read data[].image_url; for asynchronous requests, retain the task_id to query task status or receive callbacks.
Generate a blue ceramic vase on an ivory background, with soft side lighting, clean negative space, and no text; output at 1:1 and 2K.
Confirm the subject shape, background color, shadows, and final pixel dimensions. When changing colors, use the result image as the input for the next editing round.
It supports 1K, 2K, and 4K. Choose the size based on the intended use; there is no need to use the highest resolution from the start for concept drafts.
Yes. Use action=edit, image_urls, and modification instructions.
No. The default model is still nano-banana; when using 2.1, explicitly provide the full ID.
Not currently. Please use nano-banana-2.1.
Pricing varies by plan; refer to the live pricing page. Synchronous results are in data[].image_url, while asynchronous requests use the task ID for queries.
Choose a development approach or application suited to this model and follow its setup guide.
Python SDK Integration Guide
SDK installation, authentication and examplesTypeScript SDK Integration Guide
SDK installation, authentication and examplesThis is only a basic call example. See the full docs for more parameters and advanced usage.
View Docs