The world's top generative AI image models.

Nano Bananas Pro by Google is one of the most popular and powerful image generation models available in the market. You can provide multiple input images for reference and ask it to combine the images together to create unique scenes and compositions. It is also great for prompt-to-editing where you provide it with the image and then request it to edit features in the image such as changing colors, adding or removing elements, changing camera angles and much more.

Nano Bananas Pro is Google's premier generative image model and offers design teams the ability to easily combine and edit images with prompts. Select an image on the canvas, choose the Nano Bananas from the generator tray and give it a prompt. You can request changes to colors, textures, scene compositions, change actor poses, clothing and more.

To access the Generator Tray where you enter in new prompts, look on the top toolbar when you are working on a canvas project. Click on the StyleForge name (with the down chevron icon next to it). This opens up the Generator Tray where you can: A) Pick the model you want and B) enter in the prompt request and C) define any advanced settings or styles you might want to use. You can change the default image model (Flux Ultra) and use any of the available models you prefer. Note that some models require you to have an image selected on the canvas. The selected image will be used as the reference input for the model. Some models like Nano Bananas and Flux Kontext Max support two images as the input. You can select two images on the canvas, group them together, then select the Nano Bananas or Flux Kontext Max and both images will be sent to the model. You can then enter in a prompt and request the images be combined together or used however you envision. The new image(s) will be added to the canvas project. In some cases a new image is placed next to the selected image. In other cases, the selected image maybe replaced with the new generation.

Flux 2 Pro is both a text-to-image and an image-to-image generator meaning you can enter in just a prompt and get a generation, or provide up to 8 reference images at a time and request the model to work with these images to return back to you the concept you want.

We offer both image and video generators. Here is a listing of the current models we support: Ideogram 3 (text-to-image) Ideogram 3 Remix (image-to-image) Flux Ultra (text-to-image) Flux Raw (text-to-image - makes highly realistic natural images) DALL-E 3 (text-to-image - great for making artistic / illustration style images) Flux Kontext Max (text-to-image and requires a reference image for generation - supports up to two input images) Nano Bananas (Google's premier multi-image fusion model - requres a reference image for generation) Flux Kontext Multi-Image (image-to-image - combines multiple images together) Seedream 4 (text-to-image and image-to-image generator - returns high-resolution 4k images) We also offer multiple utility/special purpose models such as: Cartoonify (transforms reference image into a cartoon style animation) Change Haircut (changes hair styles on selected images that have headshots) Portrait Series (creates multiple image poses of a model and returns different poses) Professional Headshot (creates studio headshot images with neutral backgrounds) Restore Image (useful for taking older images and colorizing and restoring them) Text Removal (removes all text found on a product or scene)

Flux Raw creates exceptionally realistic images. It is a great choice for making scenes and actors that do not look like they were generated by an AI.

Flux Ultra is cutting-edge text-to-image model focusing on ultra-high resolution and detail. It excels at producing vibrant, sharp visuals, making it highly suitable for ad creatives, concept art, and digital marketing. Preserves precise prompt adherence for consistent results.

Flux Kontext Max gives design teams a powerful way to change details in existing images. You can prompt the model to change backgrounds in a scene, change actor clothing, combine elements together and more.

Flux Kontext Max is a premium text-based image editor that makes local edits to images using natural language prompts. Supports reference images for targeted changes, maintains character consistency, style binding, detailed text manipulation, and multi-turn editing for iterative creative workflows.

Ideogram 3 is a state-of-the-art text-to-image model designed for crisp text rendering, high photorealism, and exceptional style control. Its unique ability to generate text accurately within images makes it ideal for marketing, posters, and product visuals where readable, stylized text is crucial.

Flux Raw specializes in creating naturally realistic outputs with soft lighting, true-to-life textures, and subdued colors. This mode is best for photo-realistic scenes like lifestyle or landscape imagery, minimizing the "AI look" for more authentic photography-like results.

Flux 2 Pro is both a text-to-image and an image-to-image generator meaning you can enter in just a prompt and get a generation, or provide up to 8 reference images at a time and request the model to work with these images to return back to you the concept you want.

Flux 2 Pro is both a text-to-image and image-to-image generator. You can open the Generator Tray and enter a prompt and receive back an image, or you can select an image on your canvas to provide this to Flux 2 Pro to use as a reference image. You may also click on the Reference Image option on the Generator Tray and upload multiple images (up to 8) to use as reference.

Ideogram 3 Remix is an image-to-image tool that uses a reference image for style and composition guidance. Users can remix images for iterative creativity, controlling similarity levels to the original with "image weight," making it effective for both subtle edits and major creative transformations.​

DALL-E 3 is a popular AI art model by OpenAI delivers high-quality artistic generations with strong prompt adherence. Its strength is in creative transformations, multiple styles (impressionist, pop art, cyberpunk, etc.), and the ability to harmonize artistic concepts from simple to highly stylized outputs.

Seedream 4 unifies text-to-image and image-to-image generation in one system. It stands out for multi-modal creativity, 4K high-definition output, rapid inference, batch and multi-reference processing, and robust editing — ideal for advanced artistic and editing workflows in professional creative use.

Nano Banana Pro is Google’s advanced model unlocks studio-grade designs, with highly precise and localized editing, improved text rendering, and broad world knowledge. Features include multi-object blending, adjustable lighting, camera angles, color grading, and 2K–4K output. Perfect for professional use and platform-ready content.

A top-tier model tailored for vector graphic generation. It creates clean, scalable SVGs for logos, icons, and web graphics, offers accurate text integration, brand color matching, and a suite of editing tools like inpainting, erasing, and upscaling. Designed for professional design, print, and branding needs.

DALL-E 3 by OpenAI is knows for creating artistic images that lean towards vibrant colors and illustrative results. It is a great model to make impressionistic, fantasy or illustrative styled images.

DALL-E 3 by OpenAI

Seedream 4 unifies text-to-image and image-to-image generation in one system. It stands out for multi-modal creativity, 4K high-definition output, rapid inference, batch and multi-reference processing, and robust editing — ideal for advanced artistic and editing workflows in professional creative use.