Wan Image Series

POST

tasks

Wan Image Series

curl --request POST \
  --url https://api.qingbo.ai/v1/tasks \
  --header 'Content-Type: application/json' \
  --data '
{
  "model": "<string>",
  "action": "<string>",
  "prompt": "<string>",
  "negative_prompt": "<string>",
  "image_urls": [
    "<string>"
  ],
  "aspect_ratio": "<string>",
  "resolution": "<string>",
  "n": 123,
  "seed": 123,
  "watermark": true,
  "thinking_mode": true,
  "enable_sequential": true,
  "bbox_list": [
    {}
  ],
  "color_palette": [
    {}
  ]
}
'

import requests

url = "https://api.qingbo.ai/v1/tasks"

payload = {
    "model": "<string>",
    "action": "<string>",
    "prompt": "<string>",
    "negative_prompt": "<string>",
    "image_urls": ["<string>"],
    "aspect_ratio": "<string>",
    "resolution": "<string>",
    "n": 123,
    "seed": 123,
    "watermark": True,
    "thinking_mode": True,
    "enable_sequential": True,
    "bbox_list": [{}],
    "color_palette": [{}]
}
headers = {"Content-Type": "application/json"}

response = requests.post(url, json=payload, headers=headers)

print(response.text)

const options = {
  method: 'POST',
  headers: {'Content-Type': 'application/json'},
  body: JSON.stringify({
    model: '<string>',
    action: '<string>',
    prompt: '<string>',
    negative_prompt: '<string>',
    image_urls: ['<string>'],
    aspect_ratio: '<string>',
    resolution: '<string>',
    n: 123,
    seed: 123,
    watermark: true,
    thinking_mode: true,
    enable_sequential: true,
    bbox_list: [{}],
    color_palette: [{}]
  })
};

fetch('https://api.qingbo.ai/v1/tasks', options)
  .then(res => res.json())
  .then(res => console.log(res))
  .catch(err => console.error(err));

package main

import (
	"fmt"
	"strings"
	"net/http"
	"io"
)

func main() {

	url := "https://api.qingbo.ai/v1/tasks"

	payload := strings.NewReader("{\n  \"model\": \"<string>\",\n  \"action\": \"<string>\",\n  \"prompt\": \"<string>\",\n  \"negative_prompt\": \"<string>\",\n  \"image_urls\": [\n    \"<string>\"\n  ],\n  \"aspect_ratio\": \"<string>\",\n  \"resolution\": \"<string>\",\n  \"n\": 123,\n  \"seed\": 123,\n  \"watermark\": true,\n  \"thinking_mode\": true,\n  \"enable_sequential\": true,\n  \"bbox_list\": [\n    {}\n  ],\n  \"color_palette\": [\n    {}\n  ]\n}")

	req, _ := http.NewRequest("POST", url, payload)

	req.Header.Add("Content-Type", "application/json")

	res, _ := http.DefaultClient.Do(req)

	defer res.Body.Close()
	body, _ := io.ReadAll(res.Body)

	fmt.Println(string(body))

}

HttpResponse<String> response = Unirest.post("https://api.qingbo.ai/v1/tasks")
  .header("Content-Type", "application/json")
  .body("{\n  \"model\": \"<string>\",\n  \"action\": \"<string>\",\n  \"prompt\": \"<string>\",\n  \"negative_prompt\": \"<string>\",\n  \"image_urls\": [\n    \"<string>\"\n  ],\n  \"aspect_ratio\": \"<string>\",\n  \"resolution\": \"<string>\",\n  \"n\": 123,\n  \"seed\": 123,\n  \"watermark\": true,\n  \"thinking_mode\": true,\n  \"enable_sequential\": true,\n  \"bbox_list\": [\n    {}\n  ],\n  \"color_palette\": [\n    {}\n  ]\n}")
  .asString();

Wan 2.7 Image puts generation and editing in the same request: text-to-image takes only a prompt, while image-to-image, local rewrites and multi-image blending all go through edit with image_urls. This page covers the standard and pro models. Their parameter sets are identical; they differ only in the top text-to-image resolution and unit price — switching tiers means changing model and nothing else.

Quick start

curl -X POST https://api.qingbo.ai/v1/tasks \
  -H "Authorization: Bearer $WAVE_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: wan-image-demo-001" \
  -d '{
    "model": "wan2.7-image-pro",
    "action": "generate",
    "prompt": "A summer beach under a blue sky with white clouds and a distant white sailboat",
    "aspect_ratio": "16:9",
    "resolution": "4k",
    "thinking_mode": true,
    "n": 1
  }'

curl -X POST https://api.qingbo.ai/v1/tasks \
  -H "Authorization: Bearer $WAVE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "wan2.7-image",
    "action": "generate",
    "prompt": "a small fox exploring a snowy forest, storybook illustration, three scenes of the same fox",
    "enable_sequential": true,
    "n": 3,
    "resolution": "1k",
    "aspect_ratio": "1:1"
  }'

curl -X POST https://api.qingbo.ai/v1/tasks \
  -H "Authorization: Bearer $WAVE_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: wan-image-edit-001" \
  -d '{
    "model": "wan2.7-image",
    "action": "edit",
    "prompt": "Put the person from the first image into the scene from the second, matching the lighting",
    "image_urls": [
      "https://cdn.example.com/person.png",
      "https://cdn.example.com/scene.png"
    ],
    "aspect_ratio": "3:2",
    "resolution": "2k"
  }'

curl -X POST https://api.qingbo.ai/v1/tasks \
  -H "Authorization: Bearer $WAVE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "wan2.7-image",
    "action": "edit",
    "prompt": "Replace the signage inside the box with a Chinese shop name",
    "image_urls": ["https://cdn.example.com/street.png"],
    "bbox_list": [
      [[120, 80, 540, 460]]
    ],
    "resolution": "2k"
  }'

A successful submission returns a task_id. Retrieve the result with GET /v1/tasks/{task_id}, wait inline with Prefer: wait, or configure a webhook.

Request parameters

string

required

Either wan2.7-image or wan2.7-image-pro.

string

default:"generate"

generate: create an image from text; takes no input images
edit: rewrite input images; requires image_urls

Image-to-image, local rewrites and multi-image blending all use edit; they differ only in how many image_urls you send and whether you add bbox_list.

string

required

The image description or editing instruction, in Chinese or English.

string

Content to avoid in the generated result.

string[]

Input image URLs. Required for edit; up to nine images.

string

default:"1:1"

Framing: 1:1, 16:9, 9:16, 4:3, 3:4, 3:2, 2:3.

string

default:"2k"

Output resolution. wan2.7-image supports 1k and 2k; wan2.7-image-pro adds 4k for text-to-image. On both models, edit tops out at 2k.

integer

default:"1"

Number of images to generate. 1–4 in standard mode; 1–12 with enable_sequential. Billed per image, so n multiplies the unit price.

integer

Random seed. The same seed with the same parameters keeps results consistent.

boolean

Add a watermark to the output image.

boolean

default:"true"

Reasoning mode: the model thinks before generating, improving quality at the cost of longer generation time. Effective on generate only.

boolean

default:"false"

Sequential (multi-image) mode, for storyboards and series: one request returns n thematically coherent images (n up to 12). thinking_mode and color_palette have no effect when it is on; wan2.7-image-pro tops out at 2k in this mode.

array

Per-input-image boxes, [[[x1, y1, x2, y2], …], …], in absolute pixel coordinates. The outer length must equal the length of image_urls, and each image takes at most two boxes.

array

A custom palette of 3 to 10 entries (8 recommended), each shaped like {"hex": "#1B4D3E", "ratio": 12.5}, with all ratio values summing to 100.00%. Standard mode only; it has no effect in sequential mode.

See Submit a task for callback_url, callback_events, Prefer: wait, Idempotency-Key and the maximum-cost header. This series does not accept quality or mask_url. Any other parameter returns 400 and is not billed.

Limits

Condition	Limit
`n`	Up to 4 images per request in standard mode; up to 12 with `enable_sequential`
`enable_sequential: true` (`wan2.7-image-pro`)	Sequential mode tops out at `2k`
`action: "generate"`	Requires a prompt and accepts no input images
`action: "edit"`	Requires input images and is limited to `2k`
Input images	Up to nine
Resolution	`1k` / `2k` on `wan2.7-image`; `wan2.7-image-pro` adds `4k` for text-to-image
`bbox_list`	Outer length equals the length of `image_urls`, at most two boxes per image
`color_palette`	3–10 entries with `ratio` summing to 100.00%; no effect in sequential mode

Pricing

Billed at a fixed price per successfully generated image, with one price per model; a request for n images is charged n times the unit price, and the hold at submission is sized by n. Rates are in price_config on GET /v1/models and in the console’s Model Market. What a single call actually cost is the cost field on the task response — an integer quota at 500,000 quota = 1 USD, not dollars. Within a model, generate and edit cost the same, and resolution, aspect ratio, input image count, thinking_mode and color_palette do not affect the price — the unit price times n is the whole cost of a call, known before it is sent. You are only billed for a successfully generated image. Failed and cancelled tasks, and tasks that return no usable image, are refunded in full.

Response

Completed task

{
  "task_id": "task-wave1789004688b950049238",
  "model": "wan2.7-image",
  "action": "edit",
  "status": "completed",
  "progress": "100%",
  "created_at": 1788754231,
  "completed_at": 1788754279,
  "result": {
    "images": [
      {
        "expires_at": 1789089508,
        "url": ["https://cdn.example.com/result.jpg"]
      }
    ]
  },
  "urls": {
    "get": "https://api.qingbo.ai/v1/tasks/task-wave1789004688b950049238",
    "cancel": "https://api.qingbo.ai/v1/tasks/task-wave1789004688b950049238/cancel"
  },
  "billing_status": "settled",
  "cost": 13500
}

Generated images come back in result.images, and expires_at is when that link stops working. See Query Task Status for the full field list.

Available models

Model ID	Text-to-image resolutions	Editing resolutions
`wan2.7-image`	`1k` `2k`	`1k` `2k`
`wan2.7-image-pro`	`1k` `2k` `4k`	`1k` `2k`

Both support generate and edit, return up to 4 images per request (12 in sequential mode), accept up to nine input images, and share the same aspect ratios: 1:1, 16:9, 9:16, 4:3, 3:4, 3:2, 2:3.

Wan Image Series

curl --request POST \
  --url https://api.qingbo.ai/v1/tasks \
  --header 'Content-Type: application/json' \
  --data '
{
  "model": "<string>",
  "action": "<string>",
  "prompt": "<string>",
  "negative_prompt": "<string>",
  "image_urls": [
    "<string>"
  ],
  "aspect_ratio": "<string>",
  "resolution": "<string>",
  "n": 123,
  "seed": 123,
  "watermark": true,
  "thinking_mode": true,
  "enable_sequential": true,
  "bbox_list": [
    {}
  ],
  "color_palette": [
    {}
  ]
}
'

import requests

url = "https://api.qingbo.ai/v1/tasks"

payload = {
    "model": "<string>",
    "action": "<string>",
    "prompt": "<string>",
    "negative_prompt": "<string>",
    "image_urls": ["<string>"],
    "aspect_ratio": "<string>",
    "resolution": "<string>",
    "n": 123,
    "seed": 123,
    "watermark": True,
    "thinking_mode": True,
    "enable_sequential": True,
    "bbox_list": [{}],
    "color_palette": [{}]
}
headers = {"Content-Type": "application/json"}

response = requests.post(url, json=payload, headers=headers)

print(response.text)

const options = {
  method: 'POST',
  headers: {'Content-Type': 'application/json'},
  body: JSON.stringify({
    model: '<string>',
    action: '<string>',
    prompt: '<string>',
    negative_prompt: '<string>',
    image_urls: ['<string>'],
    aspect_ratio: '<string>',
    resolution: '<string>',
    n: 123,
    seed: 123,
    watermark: true,
    thinking_mode: true,
    enable_sequential: true,
    bbox_list: [{}],
    color_palette: [{}]
  })
};

fetch('https://api.qingbo.ai/v1/tasks', options)
  .then(res => res.json())
  .then(res => console.log(res))
  .catch(err => console.error(err));

package main

import (
	"fmt"
	"strings"
	"net/http"
	"io"
)

func main() {

	url := "https://api.qingbo.ai/v1/tasks"

	payload := strings.NewReader("{\n  \"model\": \"<string>\",\n  \"action\": \"<string>\",\n  \"prompt\": \"<string>\",\n  \"negative_prompt\": \"<string>\",\n  \"image_urls\": [\n    \"<string>\"\n  ],\n  \"aspect_ratio\": \"<string>\",\n  \"resolution\": \"<string>\",\n  \"n\": 123,\n  \"seed\": 123,\n  \"watermark\": true,\n  \"thinking_mode\": true,\n  \"enable_sequential\": true,\n  \"bbox_list\": [\n    {}\n  ],\n  \"color_palette\": [\n    {}\n  ]\n}")

	req, _ := http.NewRequest("POST", url, payload)

	req.Header.Add("Content-Type", "application/json")

	res, _ := http.DefaultClient.Do(req)

	defer res.Body.Close()
	body, _ := io.ReadAll(res.Body)

	fmt.Println(string(body))

}

HttpResponse<String> response = Unirest.post("https://api.qingbo.ai/v1/tasks")
  .header("Content-Type", "application/json")
  .body("{\n  \"model\": \"<string>\",\n  \"action\": \"<string>\",\n  \"prompt\": \"<string>\",\n  \"negative_prompt\": \"<string>\",\n  \"image_urls\": [\n    \"<string>\"\n  ],\n  \"aspect_ratio\": \"<string>\",\n  \"resolution\": \"<string>\",\n  \"n\": 123,\n  \"seed\": 123,\n  \"watermark\": true,\n  \"thinking_mode\": true,\n  \"enable_sequential\": true,\n  \"bbox_list\": [\n    {}\n  ],\n  \"color_palette\": [\n    {}\n  ]\n}")
  .asString();

Text Generation

Image Generation

Video Generation

Audio

Tools & Embeddings

Task Management

Account & Usage

Quick start

Request parameters

Limits

Pricing

Response

Available models

​Quick start

​Request parameters

​Limits

​Pricing

​Response

​Available models

​Related

Quick start

Request parameters

Limits

Pricing

Response

Available models

Related