A matte butter-yellow ceramic desk lamp on a warm travertine pedestal, pale sage studio backdrop, soft afternoon sunlight, realistic ceramic texture, crisp commercial product photography, wide 16:9 composition, no text, logos, or watermarks.
Nano Banana 2 Lite Text to image API
Create 1K images from text or guide an edit with up to 6 reference images.
The artwork below illustrates prompt ideas; it is not a recorded Nano Banana 2 Lite output.
Pricing
Image pricing
- 1K image
- $0.015
Playground Examples
Try production-style prompt starters.
A quiet reading corner beside a large arched window, sculptural cream armchair, oak side table and a single green branch in a ceramic vase, warm morning sunlight and long soft shadows, architectural editorial photography, wide 16:9 composition, no text, logos, or watermarks.
A tiny fox tending a rooftop garden above a sleepy seaside town, terracotta pots, lemon trees and paper lanterns, gentle gouache illustration with textured paper and warm pastel colors, wide 16:9 composition, no text, logos, or watermarks.
Similar Models
Compare adjacent capabilities before switching models.
Black Forest Labs · Image
FLUX.2 Pro
Black Forest LabsHigh-fidelity image generation and reference-guided editing with photorealistic detail, typography, and flexible framing.
OnlineTry model
Midjourney · Image
Midjourney
MidjourneyCreate expressive images with Midjourney 8.2, 8.1, 7, or Niji 7, with controls for composition, stylization, and repeating patterns.
OnlineTry model
ByteDance · Image
Seedream V5 Pro
ByteDanceByteDance image generation and editing for cinematic visuals, product shots, precise local edits, and reference-guided creative work.
OnlineTry model
Model overview
Nano Banana 2 Lite for image generation and editing
Turn a written brief into a 1K image for product concepts, social visuals, and illustrated scenes. Choose text-to-image for a new composition, or image edit when you have source images to guide the result.
Describe the scene in natural language
Build a product concept by naming the subject, materials, background, and lighting. Give the image a clear visual focus, then refine one detail at a time.
- Supports Text to image and reference-asset generation workflows.
- Fits product assets, ad previews, visual direction, and social content tests.
- Adjust configured inputs, media, and switches before generation.
Guide an edit with reference images
Add 1–6 PNG, JPEG, or WebP images and explain which features to preserve. Use references for subject appearance, setting, color, or composition, and describe the intended change explicitly.
Prompt-led generation
Choose the frame for your composition
Both generation modes support 1:1, 2:3, 3:2, 3:4, 4:3, 4:5, 5:4, 9:16, 16:9, 21:9. Choose a landscape, square, portrait, or ultrawide frame to match your composition.
- Compare available models from Alibaba, Black Forest Labs, Google, MiniMax, Midjourney, OpenAI, ByteDance, Kuaishou, SJolt AI, Suno, xAI in one place.
- Copy the request body directly to the server to reduce frontend/backend parameter drift.
- Failure states, retries, result preview, and review checkpoints stay in the same workflow.
Model characteristics
A focused workflow from prompt to image
1K output
Each task produces one 1K image. Pixel dimensions follow the selected aspect ratio.
Write a precise brief
State the subject and setting first, then add composition, light, materials, and style. For an edit, distinguish the elements to keep from those to change.
Choose a generation mode
Use text-to-image without references, or image edit with 1–6 reference images. Both modes require a prompt.
Integrate through the API
Submit the selected generation task, retain its task ID, and poll for completion or receive a webhook. Completed tasks return the image URL.
FAQ
These are the first questions to answer when evaluating this model.
Which resolution does Nano Banana 2 Lite produce?
Nano Banana 2 Lite produces 1K images. The API defaults resolution to 1K.
How do I use reference images?
Choose image edit and supply 1–6 PNG, JPEG, or WebP images with a prompt describing the desired edit. API references must use public HTTP(S) URLs.
Which aspect ratios are supported?
Both modes support 10 aspect ratios: 1:1, 2:3, 3:2, 3:4, 4:3, 4:5, 5:4, 9:16, 16:9, 21:9. The default is 16:9.