✨ Now live — VIDU Q3 Pro, our most advanced video model yet. More models coming soon.
Text-to-Image · Image-to-Image

Wan 2.5 Image

Turn a text prompt into a high-quality image, or upload an image and transform it with a prompt while keeping the original layout and subject intact. Wan 2.5 has relaxed content filters and supports negative prompts for fine creative control.

About the model

Wan 2.5 Image

Wan 2.5 Image is a versatile image model with two modes. In text-to-image, it renders a high-quality image from your prompt across a range of aspect ratios, with optional negative prompts. In image-to-image, you upload a source image and your prompt describes the edit — it preserves layout and subject structure while applying color, lighting, and stylistic changes. With relaxed filters, it offers more creative freedom (18+).

Fast and flexible. Most images generate in just a few seconds, with simple, predictable per-image pricing.
Why choose this?
Text & image to image

Create from a prompt, or edit an image you upload using a prompt.

Relaxed filters

More creative freedom — may allow NSFW / adult content (18+).

Negative prompts

Steer the result by describing what to exclude.

Layout-preserving edits

Image-to-image keeps your composition while updating the look.

Wan 2.5 Image Pricing

ModeOutputCredits
Text to Image1 image6
Image to Image (Edit)1 image7

One image per generation. Credits are only charged when a generation starts.

Choose your credit package

$3.00
300 credits
$0.0100 per credit
Buy Credits
$100.00
10,300 credits
$0.0097 per credit +300 bonus
Buy Credits
$400.00
41,500 credits
$0.0096 per credit +1,500 bonus
Buy Credits

Transparent, pay-as-you-go

Bigger top-ups earn bonus credits. No subscriptions, no hidden fees — credits never expire.

Ready to create with Wan 2.5 Image?

Sign up, grab credits, and generate your first image in seconds — no subscription.