Complex scenes, rendered precisely
Imagen 4 Ultra is Google's flagship image generation model. It handles multi-subject compositions, legible in-image text, and fine spatial detail better than most alternatives.
Specifications
What Imagen 4 Ultra delivers
16 credits per image
Higher cost reflects higher compute. Each generation produces a single output image.
Text-to-image only
Imagen 4 Ultra supports text-to-image generation. Image-to-image editing is not available for this model.
Multiple aspect ratios
1:1, 16:9, and 9:16 supported. Resolution and quality are model-determined.
Strengths
Where Imagen 4 Ultra shines
Choose this model when your prompt requires precision that simpler models struggle with.
Multi-subject compositions
Reliably places multiple distinct subjects with correct spatial relationships, reducing artifacts and blending.
In-image text rendering
Generates legible text inside the image — useful for signs, labels, UI mockups, and branded assets.
Fine detail fidelity
Handles intricate textures, patterns, and small objects with less distortion than budget models.
Prompt adherence
Closely follows complex, multi-clause prompts. Less prone to ignoring specific instructions.
Prompt directions
Looks to try with Imagen 4 Ultra
Six real assets tailored for complex compositions, text-heavy visuals, and detail-sensitive image prompts.

Watercolor garden scene
Text → Image · Painterly scene

Mountain at golden hour
Text → Image · Landscape look

Bookstore window with sign
Text → Image · Legible text

Kyoto travel poster
Text → Image · Poster typography

Sushi counter scene
Text → Image · Multi-subject scene

Skincare labels lineup
Text → Image · Product labels
FAQ
Imagen 4 Ultra FAQ
Try Imagen 4 Ultra
Sign in, select Imagen 4 Ultra, and generate your first image.
