Skip to content
Vellria
Wan 2.7 Uncensored cover art

Wan 2.7 Uncensored

Wan 2.7 Uncensored is Alibaba's image model, hosted on Vellria with nothing to install: text to image and image to image, output up to 2K. Run it in Studio or over the REST API, 8 credits (100 credits = $1.00).

ImageText to image · Image to imageAlibaba8 credits
vellria-wan-2-7
Docs

Wan 2.7 is developed by Alibaba. Vellria provides access to the model; it does not own it. All trademarks belong to their respective owners.

Input

Form

0 / 5000

Reference

0–9 images. You can upload files or paste https URLs; the provider downloads the image from the URL.

Settings

Leave empty for a random value between 0 and 2147483647.

Sign in to run

No generation yet

No generation yet

Output

No generation yet
Wan 2.7 Uncensored cover art

Cover art

8 credits

Sign in to run

Your request will cost 8 credits.

About Wan 2.7 Image

Also known as: Wan 2.7 · Alibaba Wan 2.7 Image

Wan 2.7 Image is the still-image side of Alibaba's Wan family — the same family whose open Wan 2.1 and Wan 2.2 video releases became the go-to for local NSFW video. Wan 2.7 is the newer generation, following Wan 2.5 and Wan 2.6, and this page runs its text-to-image and image-edit modes. If you searched for Wan 2.2 NSFW because you want the Wan look, this is a newer member of that family rather than the 2.2 checkpoint itself.

Fresh images come out clean, detailed and close to the wording, in a square frame. The square is a feature for avatars, profile art and grid posts; when you need a different shape, start from a photo instead, because in edit mode Wan 2.7 keeps the proportions of the first picture you attach.

That edit mode is where Wan 2.7 stands out. You can combine several photos into one scene — a person from one picture, a setting from another, an outfit from a third — and the output inherits the proportions of whichever photo you attach first. It suits composites and reworks of images you already own, where cropping to a preset shape would cut off what matters.

When the picture should move, take it to Wan 3.0, the current Wan video model, and use it as the opening frame. For literal text-to-image control, Qwen Image Edit 3.0 is the sibling to compare; for very large outputs, Seedream 5.0 Lite.

Wan 2.7 Image in action

Wan 2.7 Image outputs with the exact prompts behind them.

  • Portrait of a Korean man in his early thirties with tousled hair and a black turtleneck, neon window light casting pink and blue across his face, moody nighttime atmosphere, realistic photography

    Text to image

    Portrait of a Korean man in his early thirties with tousled hair and a black turtleneck, neon window light casting pink and blue across his face, moody nighttime atmosphere, realistic photography

    Try this prompt
  • Retro glamour shot of a brunette woman in her thirties in a vintage 1950s strapless swimsuit and cat-eye sunglasses, lounging on a striped deck chair beside a pastel pool, Kodachrome colors, no signs, no text

    Text to image

    Retro glamour shot of a brunette woman in her thirties in a vintage 1950s strapless swimsuit and cat-eye sunglasses, lounging on a striped deck chair beside a pastel pool, Kodachrome colors, no signs, no text

    Try this prompt
  • Steampunk inventor woman in her thirties with brass goggles pushed up on her head, a leather corset over a white blouse, surrounded by gears and glowing vacuum tubes in a cluttered workshop, warm light, detailed digital art

    Text to image

    Steampunk inventor woman in her thirties with brass goggles pushed up on her head, a leather corset over a white blouse, surrounded by gears and glowing vacuum tubes in a cluttered workshop, warm light, detailed digital art

    Try this prompt
  • Overhead shot of a woman in her thirties in a white satin slip dress lying on a bed of red rose petals, hair fanned out around her, soft morning light, dreamy romantic film still

    Text to image

    Overhead shot of a woman in her thirties in a white satin slip dress lying on a bed of red rose petals, hair fanned out around her, soft morning light, dreamy romantic film still

    Try this prompt
  • Art nouveau illustration of an elegant woman in her thirties with flowing auburn hair intertwined with lilies, ornate golden border, muted pastel palette, decorative poster style, no text

    Text to image

    Art nouveau illustration of an elegant woman in her thirties with flowing auburn hair intertwined with lilies, ornate golden border, muted pastel palette, decorative poster style, no text

    Try this prompt
  • 3D clay-style render of a cheerful robot barista pouring latte art in a pastel cafe, rounded shapes, soft studio lighting, playful miniature scene, no text, no signs

    Text to image

    3D clay-style render of a cheerful robot barista pouring latte art in a pastel cafe, rounded shapes, soft studio lighting, playful miniature scene, no text, no signs

    Try this prompt
  • A woman in her thirties standing behind a backlit sheer white curtain, seen only as a soft shadow silhouette through the fabric as she stretches her arms overhead at dawn, warm amber light, artistic and suggestive

    Text to image

    A woman in her thirties standing behind a backlit sheer white curtain, seen only as a soft shadow silhouette through the fabric as she stretches her arms overhead at dawn, warm amber light, artistic and suggestive

    Try this prompt
  • Before: Change his hair to platinum blond and the lighting to warm golden-hour sunlight; keep everything else the sameBefore
    After: Change his hair to platinum blond and the lighting to warm golden-hour sunlight; keep everything else the sameAfter

    Edit

    Change his hair to platinum blond and the lighting to warm golden-hour sunlight; keep everything else the same

    Try this prompt

When to pick Wan 2.7 Image

Pick it when

  • You want square avatars, profile art or grid posts with clean surface detail on skin, fabric and water.
  • You are reworking a photo you own and the output must keep that photo's original proportions instead of a preset crop.
  • You want to combine a person, a setting and an outfit from separate pictures into one composite scene.

Pick another model when

  • You need a wide or tall frame from a text prompt alone; MiniMax H3 Image offers several shapes at one price.
  • You want strictly literal text-to-image control over every listed element; compare Qwen Image Edit 3.0 on the same prompt.
  • You need very large output sizes for print or wide displays; Seedream 5.0 Lite is built for very large outputs.

Prompt or reference

Start fresh for a square image at high resolution. Attach photos to edit or merge them; the result keeps the shape of the first picture.

How to use Wan 2.7 Image on Vellria

  1. Prompt: up to 5,000 characters in the panel above.
  2. To edit or combine pictures, attach 1–9 reference images; leave them out to start from text.
  3. Cost: 8 credits — see pricing.

Example prompts

  • Square portrait of an adult woman with freckles and wet hair after a swim, bare shoulders, towel around her neck, overcast seaside light, natural skin texture.

    Try this prompt

  • Combine: the man from picture 1, the cabin interior from picture 2; he sits shirtless by the fireplace reading, warm firelight, keep the cabin's framing.

    Try this prompt

  • Square avatar of an original adult cyberpunk character, mesh top, glowing tattoos, rain and neon signs behind her, sharp focus.

    Try this prompt

  • Edit: keep the photo's framing; change the lingerie to a red silk set and add a soft rim light from behind.

    Try this prompt

  • Adult couple sharing a bath in a clawfoot tub, rose petals, candles along the tiles, top-down square composition, painterly realism.

    Try this prompt

Tips

  • Fresh images are square; start from a photo whenever you need a wide or tall frame.
  • In edit mode, attach the picture whose shape you want to keep first.
  • When merging photos, say which picture supplies which element, in plain words.
  • Keep fresh prompts specific about texture (skin, fabric, water); Wan 2.7 renders surface detail well when asked.

Specifications

Size (Text to image):2048×2048 (1:1, 2K)
Size (Image to image):2K · Matches input image
Reference images:1–9 reference images

See also: uncensored AI image generator · uncensored image editor.

Pricing

Text to image

SizeCredits
2048×2048 (1:1, 2K)8 credits

Image to image

SizeCredits
2K · Matches input image8 credits

Wan 2.7 Image vs Qwen Image Edit 3.0, Z-Image Turbo, and Wan 3.0

Wan 2.7 Image is the still-image member of Alibaba's Wan family on Vellria. It works at a single high-resolution setting and can merge several photos while keeping the first photo's shape, which sets it apart from the other Alibaba image models here.

ModelPriceMax sizeReference images
Wan 2.7 Uncensored This page8 credits2KUp to 9
Qwen 3 Uncensored8 credits2048×2048Up to 3
Z-Image Turbo Uncensored2 credits2048×2048None

Prices and limits come from the live model catalog.

Wan 2.7 Image or Qwen Image Edit 3.0?

  • Max size: 2048×2048 (Wan 2.7 Image: 2K)
  • Reference images: Up to 3 (Wan 2.7 Image: Up to 9)
  • Extra reference images: +1 credit per extra image (Wan 2.7 Image: No extra charge)
  • Prompt length: 8,000 characters (Wan 2.7 Image: 5,000 characters)
  • Settings: Seed, Negative prompt (Wan 2.7 Image: Seed)

Qwen Image Edit 3.0 offers a choice of shapes from a text prompt and reads wording literally. Pick it for a wide or tall frame written from scratch; pick Wan 2.7 when the job is combining photos you already have.

Wan 2.7 Image or Z-Image Turbo?

  • Price: 2 credits (Wan 2.7 Image: 8 credits)
  • Max size: 2048×2048 (Wan 2.7 Image: 2K)
  • Reference images: None (Wan 2.7 Image: Up to 9)
  • Extra reference images: — (Wan 2.7 Image: No extra charge)
  • Prompt length: 10,000 characters (Wan 2.7 Image: 5,000 characters)

Z-Image Turbo is the low-cost, text-only option: no reference pictures, several shapes, a lower price per image. Use it to sketch ideas, then come back here to build the composite.

Wan 2.7 Image or Wan 3.0?

Video · from 15 credits / s

Wan 3.0 is the video side of the same family. A still made here can open a Wan 3.0 clip as its first frame, which keeps the look consistent from image to motion.

Wan 2.7 Image API: price and a first request

Wan 2.7 Image costs 8 credits over the API, the same rate as in Studio. Model id vellria-wan-2-7 on POST /v1/generate/image.

curl https://vellria.com/v1/generate/image \
  -H "Authorization: Bearer vll_..." \
  -H "Content-Type: application/json" \
  -d '{
    "model": "vellria-wan-2-7",
    "prompt": "…",
    "size": "2K"
  }'

Parameters, polling and uploads: API documentation · NSFW image API.

FAQ

Does Wan 2.7 allow NSFW?

Yes, for fictional adult content. Nudity and explicit scenes between adults are allowed. Minors are refused automatically and real people without consent are prohibited.

Is Wan 2.7 Image free?

No, there is no free plan, but a new account created with Google gets 20 credits as a gift, valid for 7 days: enough to try Wan 2.7 Image, which costs 8 credits. After that every run is paid from prepaid credits at the rate this page shows before you start, and a run that fails at the provider is refunded automatically. Is there a free uncensored AI generator? covers the cheapest way to a first result.

Is this Wan 2.2?

No. It is Wan 2.7 from Alibaba, a newer generation of the Wan family. Wan 2.2 is the earlier open video release and runs here as Wan 2.2; for video with sound use Wan 3.0.

Why are my fresh Wan 2.7 images square?

Text-to-image on this model is square by design. Attach a photo and the result keeps that photo's shape instead.

Can Wan 2.7 merge several photos?

Yes. Attach several and describe how they combine; the limit is listed in the specifications table.

Related models

Turn a still into a clip with one of our video models.

Wan 3.0 Pro Uncensored cover art

Wan 3.0 Pro Uncensored

Step up for final takes: 2–30 s with sound up to 1080p; start from a still to keep your framing.

Video

from 20 credits / s
MiniMax H3 Video Uncensored cover art

MiniMax H3 Video Uncensored

Test shot ideas before the final take: 4–15 s clips up to 4K, or see how a still moves.

Video

from 6 credits / s
MiniMax H3 Max Uncensored cover art

MiniMax H3 Max Uncensored

For the take you keep: 5–15 s clips up to 4K with sound; start from a still if you have one.

Video

from 15 credits / s