Google text-to-image model
Imagen 4 AI image generator
Imagen 4 on RenderFlow AI is Google's photoreal text-to-image model at 5 credits per image. The canvas adds aspect ratio, 1k or 2k resolution, a negative prompt, a seed, and 1 to 4 images. It does not edit an uploaded picture.
- Text to image
- 1k or 2k
- 5 aspect ratios
- 5 credits per image.
Controls and the credit price match the RenderFlow AI catalog as checked on 7 November 2026.
Generate an Imagen 4 image
Write the subject, the light, and the camera before you add style words. The credit total appears before you run the model.
Your generated image will appear here.
Enter a prompt and click 'Generate Image' — or start from an example:
Examples
Imagen 4 examples from real prompts
These six images are Imagen 4 outputs on RenderFlow AI, covering a greenhouse, a portrait, a product, a landscape, a car, and a macro frame. Open one to copy the prompt, then change the subject.
Environment concept
Golden-hour greenhouse
Prompt: Abandoned Victorian greenhouse overgrown with tropical plants, shafts of golden light through broken glass ceiling, moody atmosphere
Professional headshot
Executive portrait
Prompt: Professional headshot of a confident woman CEO in a modern office, natural window light, shallow depth of field, corporate photography
E-commerce photography
Artisan leather product shot
Prompt: Artisan leather bag product shot on aged wooden table, warm directional light, minimalist composition, e-commerce photography
Landscape photography
Misty mountain valley
Prompt: Misty mountain valley at dawn with a winding river, layers of fog between peaks, drone photography, epic landscape
Automotive campaign
Neon electric sports car
Prompt: Sleek electric sports car in a neon-lit underground parking garage, reflections on wet concrete, automotive photography
Macro photography
Rainbow dewdrop
Prompt: Macro photograph of a dewdrop on a spider web catching rainbow light, extreme close-up, nature photography
What Imagen 4 is for on RenderFlow AI
The catalog describes Imagen 4 as Google image generation for photorealistic, clean commercial imagery with strong prompt fidelity.
Product shots, brand stills, and straightforward photoreal prompts are the jobs it lists.
The author on the model record is Google, and the endpoint is google/imagen-4.
Google Imagen 4 next to Imagen 3 and Ultra
Imagen 3 also costs 5 credits per image and uses the same five aspect ratios, but its canvas is prompt, aspect ratio, and seed. Imagen 4 adds resolution, a negative prompt, and a batch of 1 to 4 images at the same per-image price. Imagen 4 Ultra costs 8 credits per image and exposes aspect ratio and seed only. Use Ultra when you already like the idea and want that higher-priced endpoint, not because this page is missing a hidden quality slider.
Google Image 4 when the brief is a photograph
Describe the object, the light direction, the lens or crop, and what must stay out of frame. Imagen 4 is a generator, not an editor. If you already have the product photo and need to change it, use an image editor such as Qwen Image Edit. If the deliverable is a poster with several lines of type, start with Qwen Image or GPT Image 2 and keep Imagen 4 for the photographic plate.
How to run an Imagen 4 brief
Set the frame first, spend 1k credits on the composition, then pay for 2k only on the keeper.
Imagen 4 Google product prompt
Artisan leather shoulder bag on an oak table, warm side light from the left, soft shadow, 3/4 view, empty background, product photograph. No logo, no text.
Start at 1:1 or 4:3 and 1k. Add the negative prompt "logo, text, extra straps" if hardware multiplies. This is a prompt to adapt, not a new output beyond the gallery above.
Google Imagen 4 portrait prompt
Professional headshot of a confident woman in a modern office, window light from camera left, shallow depth of field, natural skin texture, looking at the camera.
Lock the seed once the face works, then change only the wardrobe or background. This is a prompt to adapt, not a new output beyond the gallery above.
Google Image 4 landscape prompt
Misty mountain valley at dawn, winding river, layered fog, aerial photograph, natural color, no people, no text.
Use 16:9 for a hero and 9:16 only when the placement is vertical. Move to 2k after the composition is right. This is a prompt to adapt, not a new output beyond the gallery above.
-
Match the Imagen 4 frame to the placement
Choose one of 1:1, 16:9, 9:16, 4:3, 3:4. Square for a tile, 16:9 for a banner, 9:16 for a story, 4:3 or 3:4 for a feed. Generating the wrong ratio and cropping later throws away the composition you just paid for.
-
Stay on Imagen 4 at 1k until the picture is right
Resolution options are 1k or 2k. The default is 1k. Use 1k while you compare 1 to 4 images. Switch to 2k for the file you will actually place. The catalog price is 5 credits per image. Confirm the batch total in the canvas, because a four-image run is four images.
-
Use the Imagen 4 negative prompt and seed as brakes
The negative prompt is optional. Name the defects you keep seeing, such as extra fingers, logos, or text. The seed is optional. Set it when you want the next try to stay close, and clear it when you want a new composition.
Imagen 4 settings and limits
The table follows the live Imagen 4 schema on RenderFlow AI. It is not a Google API feature list.
| Prompt | Required. This page does not upload a source image. |
|---|---|
| Aspect ratio | 1:1, 16:9, 9:16, 4:3, 3:4. Default 1:1. |
| Resolution | 1k or 2k. Default 1k. No 4k option on this endpoint. |
| Image count | 1 to 4, default 1. |
| Negative prompt | Optional. |
| Seed | Optional integer. |
| Credits | 5 credits per image. Pricing explains how credits are spent. Check the canvas total before a multi-image run. |
Imagen 4 questions
What is the difference between Google Imagen 4 and Imagen 4 Ultra?
Google Imagen 4 costs 5 credits per image and adds a 1k or 2k resolution control, a negative prompt, and a count of 1 to 4 images. Imagen 4 Ultra costs 8 credits per image. Its canvas has aspect ratio and seed only. Draft on this page, then move an approved idea to Ultra if you want that higher-priced endpoint.
Which aspect ratios and resolutions does Google Image 4 expose here?
The Imagen 4 canvas offers 1:1, 16:9, 9:16, 4:3, and 3:4. Resolution is 1k or 2k, defaulting to 1k. Those are the control labels. This page does not add a separate pixel-size field or a 4k option.
How does Imagen 4 compare with Imagen 3?
Both cost 5 credits per image and share the same five aspect ratios. Imagen 3 exposes prompt, aspect ratio, and seed. Imagen 4 also exposes resolution, a negative prompt, and 1 to 4 images per run. Pick Imagen 3 when an existing workflow already uses that endpoint. Pick Imagen 4 when you want the extra controls.
When should an Imagen 4 Google prompt leave this page?
Leave when the job is an edit of an existing file, a long text layout, or a size this canvas does not list. Qwen Image and GPT Image 2 are the text-layout neighbors. Imagen 4 does not accept a source image.
Can Imagen 4 render text in images?
Do not use Imagen 4 as a typesetting tool. If words must be exact, proofread them at full size or generate the photograph here and set the type in a layout tool. Qwen Image and GPT Image 2 are the better first stop for posters and labels.
Can I use Imagen 4 images commercially?
RenderFlow AI does not grant a special commercial license on this page. Review the current plan terms and Google's usage terms for Imagen before you publish client work, ads, or products.
Open Imagen 4 in the image generator
Run the same prompt at 1k, keep the seed on the best frame, and only then switch that frame to 2k or to Imagen 4 Ultra.
Open Imagen 4More models beside Imagen 4
Imagen 4 Ultra
Use Google Imagen 4 Ultra for premium photorealistic AI image generation.
Google Imagen 3
Create photorealistic images with Google Imagen 3.
GPT-Image-2 Generator
Next-gen OpenAI image generation with three quality tiers.
Qwen Image Generator
Create images with strong text rendering and layout control.
Flux Kontext Pro
Edit images with context-aware FLUX prompt controls.
Flux Kontext Max
Use the premium FLUX Kontext image editing model.