Start with a sentence. Build a scene.
Describe what you want to see and create an image from text. Explore photographic scenes, illustrations and visual concepts in the Nano Banana workspace, then refine the result that comes closest to your idea.
Use advanced AI technology to transform your imagination into beautiful images
Sign up now to experience 4K resolution and Pro model - no credit card required
Use the Gemini model’s real-world knowledge and deep reasoning capabilities to deliver precise, detailed, rich image results. Annotate pictures, represent data as infographics, or turn handwritten notes into diagrams.
Sharp, legible text helps you create impactful posters, intricate diagrams, and detailed product mockups. Describe what type of font you want, or simulate different types of handwriting.
Precisely capture and transfer artistic styles—whether oil painting, watercolor, illustration, or photography—flawlessly replicate and apply to your creations.
Support up to 4K ultra-high resolution output with every detail crystal clear, meeting professional printing and commercial use requirements.
Text to image
An AI image generator creates a picture from your description. Your prompt supplies the subject, setting, style and composition; the model interprets them together. No source image is required, so you can begin with an idea that does not yet have a visual reference.
This page opens in Text-to-Image mode. Already have a photo to change? Use the AI image editor, or explore image to image for transformations based on a reference.
Give each prompt a clear visual direction, from the subject to the way the scene is framed.
Describe the location, subject, camera view and lighting. Try a still life beside a window, a misty mountain landscape or an imagined interior. These are generated images, not records of real events.
Ask for an editorial drawing, watercolor, paper-cut composition or another visual treatment. A specific palette and material description help communicate the look more clearly than a long list of style words.
Try different arrangements, colors or settings to develop an idea. Change one part of the prompt between attempts so you can tell which instruction improved the result.
Download a version you like, or select Continue Edit to use it as an image reference for another adjustment. Move from generating a new scene to editing its details in the same workspace.
Use the workspace above to describe, generate and review your first image.
Keep Text-to-Image selected and write what should appear. Include the main subject, its surroundings, the visual style and where you want important elements placed.
Select a model and choose the aspect ratio, resolution and output count offered for it. Sign in and check the displayed credit cost before submitting your prompt.
Review the composition, faces, hands and any lettering. Download the image, adjust your prompt for another attempt, or use Continue Edit to revise a selected result.
A useful starting structure is subject + setting + style + composition. Adapt these examples, or browse the prompt library.
Photographic scene
A ceramic teapot and two small cups on an oak table beside a large window. Soft early morning light from the left, visible steam, muted cream and olive colors. Close eye-level view with the teapot as the focal point. No lettering.
Editorial illustration
An editorial illustration of a compact city with rooftop gardens, cycle paths and a central park. Textured gouache on paper, forest green and terracotta palette. Elevated wide view, simple building shapes and a clear focal point in the park.
Concept design
A small reading corner with a curved walnut bookshelf, a linen armchair and a floor lamp. Warm evening light, realistic materials and a calm atmosphere. Wide composition with the chair on the right and open wall space on the left.
The prompt is a description of the picture you want, not a collection of unrelated keywords.
Start with who or what the image is about, then add setting, light and materials. If the scene feels crowded, reduce the number of subjects before adding more descriptive words.
Say whether the subject should be centered, off to one side, close up or seen from above. Leave empty space in the composition if you plan to add a headline later in a design tool.
Check small details and spelling instead of judging only the thumbnail. Use the image upscaler to enlarge a finished image, or the AI image editor for a focused visual change.
Choose the perfect plan for your AI image creation needs.
Cancel anytime
Perfect for trying out AI image generation and small projects.
What's included
Perfect for experiencing AI image generation and editing effects.
Ideal for creators, designers, and content professionals.
Everything in Starter
🔥 Limited Time Offer! Get Pro Now
For teams, agencies, and high-volume commercial use.
Everything in Pro
Perfect for teams and businesses
For any questions about payment, contact our support team:Contact support
Answers about text prompts, image references, credits and revisions.
It creates images from a written description using a generative model. The prompt can describe a photographic scene, an illustration or an imagined design. This page uses the Nano Banana workspace for text-to-image generation.
No. Text-to-Image mode starts from a text prompt. If an existing photo or sketch is essential to your idea, use image to image to provide that visual reference.
Generation uses credits. If your account has free credits available, you can spend them on supported tasks. The model, resolution and output count affect the cost shown before generation. See pricing for current plans.
Name the main subject, setting, style, lighting and composition. For example, “a red bicycle against a pale wall, soft afternoon light, side view, realistic photography” gives the model more direction than “a nice bicycle.”
You can request both. Describe the desired medium and visual qualities rather than relying on the model to infer them. Results depend on the selected model and prompt, and may need further attempts or edits.
You can request specific wording, but spelling, letter shapes and placement may be imperfect. Keep text requests short and inspect the output. For exact typography, generate the background first and add text in a design tool.
Select Continue Edit on a result to use it as a reference, then describe the change and what should stay. You can also revise the original text prompt and generate a new version. Save useful outputs before trying another direction.
Switch tools without breaking your creative flow.
Fast, cost-efficient 1K image generation and editing
Versatile image generation and editing with up to 4K output
Improved realism, prompt following and text rendering
Professional image creation with precise control and up to 4K output
High-quality poster design and image editing
Text-to-image, affordable and fast!
Precise AI image generation and editing with 1K/2K output
Create and edit images from prompts and references
AI image generation with near-perfect text rendering
Create and edit 1K, 2K, or 4K images with Flare or Sunburst
Explore image creation, editing, and prompt ideas
Generate and edit polished 1K or 2K images from text and references
Start with a subject, add a setting and choose a visual style. Bring the first version to life with Nano Banana.
