A starting image. A specific edit.
Give your next visual a head start. Upload a product shot, portrait or sketch, then describe the background, style or detail you want to change. Bimg AI uses your reference and prompt to generate a new version in the Nano Banana workspace below.
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.
Build on an existing image
Image to image AI, often called img2img, combines a picture with written instructions to generate a new image. The picture shows the subject and composition; your prompt sets the direction. Use it when describing a scene from scratch would leave out details you already have in a photo or drawing.
Bimg AI brings reference-based editing into the Nano Banana workspace. This page opens in image editing mode. Choose a model, add your source image and write the change you want to explore.
Start with an asset you know, then explore a different setting, illustration style or design direction.
Move a product from a plain backdrop to a seasonal display or a studio set. Specify the surface, light direction and props, and ask to retain the product's outline and label. Check those details before publishing.
Try a shared palette, background mood or illustration treatment across several source images. Reuse the style instructions and review each output for differences in color, framing and recognizable details.
Turn a rough drawing into a visual concept for a room, object or scene. Describe the surfaces and finish, and list which positions and proportions should follow the sketch.
Use Continue Edit to take a generated result into another round. Adjust one part, compare the versions and return to your original reference when you want to try a different direction.
Three steps from a source file to a version you can review and download.
Upload a JPG, PNG or WebP with the subject visible. Image to Image is selected on this page. Check the upload area for the selected model's file-size and reference-count limits.
Name the change first, then the details to keep. For example: replace the backdrop with a warm studio scene; keep the product, label and camera angle. Choose the available output settings and review the credit cost.
Sign in and generate your image. Compare it with the source, especially faces, text, product edges and shadows. Download a version you like or choose Continue Edit for a focused follow-up.
A useful editing brief states the change, what should stay and the desired finish. Replace the subject and scene in these examples with details from your own reference.
Campaign product image
Place the ceramic mug from this photo on a dark green shelf with a folded cream linen cloth beside it. Keep the mug's handle, proportions, glaze and printed design. Use soft light from the upper right and realistic shadows. Keep the original viewing angle and add no text.
Illustrated social post
Reinterpret this street photo as a clean screen-print illustration with a limited navy, coral and ivory palette. Preserve the main buildings, road direction and people placement. Use flat shapes and subtle paper texture. Leave the sky uncluttered and do not add lettering.
Design concept
Use this chair sketch as the shape reference. Render the chair in light ash wood with an oatmeal fabric seat against a neutral studio background. Keep the leg arrangement and backrest outline from the drawing. Add gentle side lighting without adding extra furniture.
Reference-based generation can reinterpret parts of the source. Build a short review into your workflow.
Look at faces, product shapes, patterns and small lettering side by side with the source. If a detail has drifted, return to the original and make the preservation instruction more specific.
A new background needs plausible light, shadows and perspective. Ask for a contact shadow beneath the subject and describe where the light comes from. Avoid combining several conflicting style instructions.
Need a transparent cutout? Use the background remover. Need a larger final file? Try the AI image upscaler after you have settled on the edit.
Choose a subscription or credit pack for Nano Banana image editing, Nano Banana Pro/2, enhancement tools, and AI video models.
Cancel anytime
Best for testing Nano Banana, image tools, and light creative projects.
What's included
Good for exploring Bimg AI image and video workflows.
Best for creators who need more Nano Banana Pro, 4K output, and AI video generation.
Everything in Starter
Best value for Nano Banana Pro and AI video generation.
Best for teams and agencies with high-volume image and video workflows.
Everything in Pro
Built for teams and high-volume Bimg AI workflows.
For any questions about payment, contact our support team:Contact Us
What to upload, how to guide an edit and what to expect from the result.
It generates a new image using an uploaded picture and a text prompt together. The reference provides visual information while the prompt describes an edit or transformation. Common uses include changing a product scene, trying an art style and developing a sketch into a more finished concept.
Use image to image when a photo, drawing or previous result already contains the subject or layout you want. Text to image is useful when you want to start with a written idea and have no reference. You can switch between these modes in the workspace.
Image generation uses credits. Available free credits on your account can be used for supported tasks; further usage may require a purchase. Check the displayed cost before generating because the model, resolution and number of outputs affect it. Current credit options are listed on the pricing page.
Yes, you can request a new setting and ask the model to retain the product's shape, color and markings. Describe the new surface and lighting as well. Generated edits can still change labels or fine geometry, so inspect the output before using it in a product listing.
Upload your photo, name the target style and explain which content should remain recognizable. For example, ask for an anime illustration while retaining the pose and clothing. You can also describe the palette, line work and background to make the direction clearer.
The workspace accepts JPG, PNG and WebP. A clear image with an unobstructed subject is usually easier to guide than a tiny, heavily compressed source. File-size limits and the number of references depend on the selected model and are shown in the upload area.
No. The model generates a new result and may reinterpret details even when you ask to preserve them. Specify the important features, make one focused change at a time and compare each version with the source. Editing is guided by your prompt; this workspace does not offer a denoising-strength slider.
Use Continue Edit to make a specific follow-up, such as softening a shadow or changing the background color. If several rounds have moved too far from the source, upload the original again and simplify the instructions. Save a version you like before exploring another direction.
Switch tools without breaking your creative flow.
Fast, cost-efficient 1K image generation and editing
Advanced Nano Banana AI image editing with 2K/4K output
Next-gen Nano Banana image editing with clearer text and sharper details
4K images, clearer text, consistent edits
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
Plan product shots and campaign images from a clear visual brief
Generate and edit polished 1K or 2K images from text and references
Choose a source photo or sketch and give it one clear editing instruction. Explore the result with Bimg AI.
