Wan 2.6 Image

Professional image editing with instant serverless access on Together AI.

About model

Wan 2.6 Image has 20B parameters. On Together AI it costs $0.03 per image.

Wan 2.6 Image is Alibaba's 20B parameter image transformation model, built for complex image-to-image workflows, multi-reference style transfer, and precise structural edits. With Together AI, creative teams and developers can leverage Wan 2.6 for high-fidelity, production-ready image editing at scale.

‍

Quickstart guides

Image

Images Docs

API usage

cURL
Python
Typescript

Endpoint:

Wan-AI/Wan2.6-image

curl -X POST "https://api.together.ai/v1/images/generations" \
  -H "Authorization: Bearer $TOGETHER_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "Wan-AI/Wan2.6-image",
    "prompt": "Draw an anime style version of this image.",
    "width": 1024,
    "height": 768,
    "steps": 28,
    "n": 1,
    "response_format": "url",
    "image_url": "https://huggingface.co/datasets/patrickvonplaten/random_img/resolve/main/yosemite.png"
  }'

from together import Together

client = Together()

imageCompletion = client.images.generate(
    model="Wan-AI/Wan2.6-image",
    width=1024,
    height=768,
    steps=28,
    prompt="Draw an anime style version of this image.",
    image_url="https://huggingface.co/datasets/patrickvonplaten/random_img/resolve/main/yosemite.png",
)

print(imageCompletion.data[0].url)

import Together from "together-ai";

const together = new Together();

async function main() {
  const response = await together.images.create({
    model: "Wan-AI/Wan2.6-image",
    width: 1024,
    height: 1024,
    steps: 28,
    prompt: "Draw an anime style version of this image.",
    image_url: "https://huggingface.co/datasets/patrickvonplaten/random_img/resolve/main/yosemite.png",
  });

  console.log(response.data[0].url);
}

main();

Model card
Architecture Overview:
• Multimodal Diffusion Transformer (MMDiT) architecture with 20 billion parameters
• Supports 1-3 reference images for multi-source composition and style transfer
• Generates 1-4 output images per request with flexible aspect ratios and HD resolutions
• Optional LLM-powered prompt expansion for enhanced detail in complex scenes

Training Methodology:
• Optimized to preserve identity, textures, and proportions from reference imagery
• Deep multilingual and cultural understanding, particularly in Asian contexts
• Trained for advanced spatial reasoning, ensuring elements follow a logical visual structure

Performance Characteristics:
• Multi-subject consistency that maintains visual coherence across characters and environments
• Precise, structure-preserving edits (e.g., changing lighting or materials without breaking the layout)
• Reliable text rendering for integrating typography naturally into images
‍
Applications & use cases
Marketing & Brand Design:
• Professional marketing materials requiring text integration and logical layout
• Brand asset creation with consistent preservation of logos and visual identity
• Product visualization and mockups with photorealistic lighting and textures
• Style transfer for campaign creative maintaining brand consistency across variations

Design & Production:
• Photo retouching and enhancement with structure-preserving edits
• Virtual staging and scene composition combining elements from multiple sources
• Character design and illustration with multi-subject consistency
• UI element generation and design iteration with precise control

Creative Workflows:
• Style exploration and artistic transformations from reference imagery
• Iterative design refinement with targeted modifications
• Cultural and regional content creation leveraging multilingual understanding
• Production-ready asset generation for diverse creative applications
‍

Model specifications

Model data

Model provider
Alibaba
Type
Image
Main use cases
Image-to-Image
Deployment
Serverless
Endpoint
Wan-AI/Wan2.6-image
Parameters
20B
Price
$0.03 / image
Input modalities
Text
Image
Output modalities
Image

Released
January 28, 2026
Category
Image

Run in Playground

Quickstart docs

Deploy Model

Wan 2.6 Image

About model

API usage

Model card

Applications & use cases