# Wan | v2.6 | Image to Image

Wan 2.6 Image-to-Image transforms input images with precise, high-quality edits while maintaining visual consistency.

## API Information

- **Model Slug:** wan-v2-6-image-to-image
- **Branded URL:** https://www.eachlabs.ai/alibaba/wan-v2-6/wan-v2-6-image-to-image
- **Provider:** Alibaba
- **Category:** Image to Image
- **Output Type:** image
- **Status:** active
- **Version:** 0.0.1
- **Base Cost:** Per image pricing: $0.03/image
- **Estimated Processing Time:** 80 seconds
- **Last Updated:** 2026-04-02
- **Interactive Demo:** https://www.eachlabs.ai/ai-models/wan-v2-6-image-to-image

## Pricing

- **Charge Type:** dynamic
- **Pricing Details:** Per image pricing: $0.03/image

### Pricing Rules

| Condition | Pricing |
| --- | --- |
| Rule 1 | Per image pricing: $0.03/image |

## Input Schema

| Parameter | Type | Required | Default | Constraints | Description |
|-----------|------|----------|---------|-------------|-------------|
| prompt | string | Yes | - | - | Text prompt describing the desired image. Supports Chinese and English. Max 2000 characters. Example: 'Generate an image using the style of image 1 and background of image 2'. |
| image_urls | array | Yes | - | 1–3 | Reference images for editing (1-3 images required). Order matters: reference as 'image 1', 'image 2', 'image 3' in prompt. Resolution: 384-5000px each dimension. Max size: 10MB each. Formats: JPEG, JPG, PNG (no alpha), BMP, WEBP. |
| negative_prompt | string | No | - | - | Content to avoid in the generated image. Max 500 characters. |
| image_size | string | No | square_hd | square_hd,square,portrait_4_3,portrait_16_9,landscape_4_3,landscape_16_9 | Output image size. Use presets like 'square_hd', 'landscape_16_9', 'portrait_9_16', or specify exact dimensions with ImageSize(width=1280, height=720). Total pixels must be between 768*768 and 1280*1280. |
| num_images | integer | No | 1 | 1–4 | Number of images to generate (1-4). Directly affects billing cost. |
| enable_prompt_expansion | boolean | No | true | - | Enable LLM prompt optimization. Significantly improves results for simple prompts but adds 3-4 seconds processing time. |
| seed | integer | No | - | - | Random seed for reproducibility (0-2147483647). Same seed produces more consistent results. |
| enable_safety_checker | boolean | No | true | - | Enable content moderation for input and output. |

## Example Request

```bash
curl -X POST https://api.eachlabs.ai/v1/prediction/ \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
  "model": "wan-v2-6-image-to-image",
  "input": {
    "prompt": "Place the princess from image 1 in the elegant living room from image 3, holding and gently examining the star-tipped magical wand from image 2. The wand emits a soft, warm golden glow that subtly illuminates her face and hands.\n\nShe is seated gracefully, with a calm, composed expression, maintaining her exact facial features, hairstyle, and proportions from the reference image. Do not alter or beautify her face in any way.\n\nThe living room feels regal yet intimate: classic furniture, soft fabrics, warm ambient lighting, and refined decorative details. The atmosphere should feel magical but grounded in reality.\n\nThe glow from the wand reflects softly on nearby surfaces, creating a delicate, enchanting mood without overpowering the scene.\n\nPhotorealistic, natural colors, no stylization, no fantasy effects beyond the glowing wand.\nNo text, no logos, no watermarks.\nUltra-detailed, realistic lighting, high-resolution.",
    "image_urls": [
      "https://storage.googleapis.com/magicpoint/inputs/wan-v2-6-image-to-image-input1.png",
      "https://storage.googleapis.com/magicpoint/inputs/wan-v2-6-image-to-image-input2.png",
      "https://storage.googleapis.com/magicpoint/inputs/wan-v2-6-image-to-image-input3.png"
    ]
  }
}'
```

## Output Schema

Response returned by `GET /v1/prediction/{id}` when the job completes:

```json
{
  "status": "success",
  "predictionID": "string",
  "output": "string (URL of generated image)",
  "metrics": {
    "predict_time": "number (seconds)"
  }
}
```

## Polling

```bash
curl https://api.eachlabs.ai/v1/prediction/{PREDICTION_ID} \
  -H "Authorization: Bearer YOUR_API_KEY"
```

| Status | Meaning |
|--------|---------|
| `processing` | Still running — poll again |
| `success` | Done — read `output` |
| `error` | Failed — read `message` / `details` |

## Webhook (alternative to polling)

Pass `"webhook_url": "https://your.host/path"` in the create request. Eachlabs POSTs this payload when the job ends:

```json
{
  "exec_id": "prediction-uuid",
  "status": "succeeded",
  "output": "https://...",
  "error": ""
}
```

`status` is `"succeeded"` or `"failed"`. `exec_id` equals the `predictionID` from create. Return 2xx within 30 seconds.

## Errors

Error body: `{ "status": "error", "message": "...", "details": "..." }`

| Code | Meaning |
|------|---------|
| `400` | Invalid input |
| `401` | Missing / invalid `Authorization` bearer token |
| `404` | Unknown model or prediction id |
| `429` | Rate limit — 100 creates / min, 10 concurrent per key |
| `5xx` | Retry with backoff |

## Overview

Wan-v2.6-image-to-image is an advanced image-to-image AI model developed by Alibaba as part of the Wan 2.6 series, specializing in editing and generating images using 1-3 reference images. It enables precise modifications such as style transfer from references, maintaining subject consistency across outputs, and creating complex compositions by combining elements from multiple inputs, producing 1-4 output images per request.

Key features include multi-image input support for sophisticated scene assembly, prompt expansion via integrated LLM optimization for enhanced detail, and flexible output resolutions up to 1280x1280 pixels. The model supports both English and Chinese prompts, with negative prompting to avoid undesired elements, making it suitable for diverse creative and technical image editing tasks.

Its underlying architecture leverages large-scale multimodal diffusion transformer (MMDiT) technology, similar to models like Qwen-Image-Edit, optimized for high-fidelity edits while preserving identity, textures, and proportions from reference images. What sets it apart is its ability to reference images explicitly in prompts (e.g., "image 1" for style, "image 2" for background), delivering consistent, professional-grade results in a single generation pass.

## Usage Notes

- API Base URL: `https://api.eachlabs.ai/v1`
- Authentication: send `Authorization: Bearer YOUR_API_KEY`. Generate a key from the Eachlabs dashboard at https://www.eachlabs.ai/dashboard/api-keys.
- File-typed parameters (`*_url`, `image_url`, `video_url`, `audio_url`, etc.) accept publicly-reachable HTTPS URLs only. Upload your asset first (GCS / S3 / your CDN) and pass the resulting URL. Data-URIs and localhost URLs are rejected.
- For structured parameters (arrays / objects) send real JSON values, not stringified payloads.
- Monetary values are reported in USD; per-token / per-megapixel rates may be billed in micro-cents internally.
- Prefer `webhook_url` over polling for long-running predictions — see the Webhook Callback section.