Use Wan 2.5 I2I Preview with Comfy Router

Call wan/wan2.5-i2i-preview through Comfy Router: endpoint, request shape and the response Router returns.

API Reference for wan/wan2.5-i2i-preview, served by Comfy Router from Wan.

Quick start

Create a key at platform.comfy.org/profile/api-keys and export it as COMFY_API_KEY. The Python and TypeScript snippets use the Comfy SDKs (pip install comfy-sdk, npm install @comfyorg/sdk); the cURL snippet is the same call over raw HTTP.

Model ID: wan/wan2.5-i2i-preview

Endpoint: POST https://api.comfy.org/v2/models/wan/wan2.5-i2i-preview

Python

from comfy_sdk import Comfy

# Reads COMFY_API_KEY from the environment. Each call sends a fresh
# Idempotency-Key and waits up to 10 minutes for the finished result.
with Comfy() as client:
    result = client.models.run(
        "wan/wan2.5-i2i-preview",
        {
            "input": {
                "images": ["https://example.invalid/red-maple-leaf.png"],
                "prompt": "Make the leaf golden.",
            },
            "parameters": {
                "n": 1,
                "size": "768*768",
            },
        },
    )

print(result)

Schema

Input

input · object · required

Enter basic information, such as prompt words, images, etc.

input.images · string[] · required

Array of image URLs for image-to-image generation

input.negative_prompt · string

Reverse prompt words to describe content that you do not want to see in the image

input.prompt · string · required

Positive prompt words to describe expected image elements and visual features. Support Chinese and English, length not exceeding 2000 characters

model · string

The ID of the model to call for image-to-image generation. NOT constrained on this component: Comfy Router fills it from the {model} path segment of POST /v2/models/wan/{model}. A direct v1 call to POST /proxy/wan/api/v1/services/aigc/image2image/image-synthesis MUST supply it, and the enum of accepted spellings lives on that operation’s own component, WanImage2ImageGenerationRequest.

parameters · object

Image processing parameters

parameters.n · integer · default: "1"

Number of generated images. Range 1-4, default is 1

Range: 1 to 4

parameters.seed · integer

Random number seed to control randomness. Range [0, 2147483647]

Range: 0 to 2147483647

parameters.size · string · default: "\"1280*1280\""

The output image resolution in the format widthheight. Default is 12801280. The API accepts a pixel area between 589824 (768768) and 1638400 (12801280) with an aspect ratio between 1:4 and 4:1

parameters.watermark · boolean · default: "false"

Whether to add watermark logo in lower right corner

Generated from the schema Router serves at GET /v2/models/wan/wan2.5-i2i-preview/openapi.json, the same document it validates a call against before the request reaches the provider.

Output

output · object · required

output.actual_prompt · string

Actual prompt after intelligent rewriting (for video tasks)

output.check_audio · string

Audio URL for I2V tasks with audio generation

output.code · string

The error code for the failed request (not returned if request is successful)

output.end_time · string

Task completion time

output.message · string

Detailed information about the failed request (not returned if request is successful)

output.orig_prompt · string

Original input prompt (for video tasks)

output.results · object[]

List of task results for image generation tasks

output.results[].actual_prompt · string

Actual prompt after intelligent rewriting (if enabled)

output.results[].code · string

Image error code (returned when some tasks fail)

output.results[].message · string

Image error information (returned when some tasks fail)

output.results[].orig_prompt · string

Original input prompt

output.results[].url · string

Generated image URL address

output.scheduled_time · string

Task execution time

output.submit_time · string

Task submission time

output.task_id · string · required

Task ID

output.task_metrics · object

Task result statistics for image generation tasks

output.task_metrics.FAILED · integer

Number of failed tasks

output.task_metrics.SUCCEEDED · integer

Number of successful tasks

output.task_metrics.TOTAL · integer

Total number of tasks

output.task_status · string · required

Task status

Possible values: PENDING, RUNNING, SUCCEEDED, FAILED, CANCELED, UNKNOWN

output.video_url · string

Video URL for completed video generation tasks. Link validity period 24 hours

request_id · string · required

Unique request identifier

usage · object

Output information statistics. Only successful results are counted

usage.SR · integer

Video resolution level (I2V and wan3.0-video tasks)

usage.duration · number

Duration of generated video in seconds (I2V and wan3.0-video tasks)

usage.fps · integer

Frame rate of the generated video (wan3.0-video tasks)

usage.image_count · integer

Number of generated images (T2I tasks)

usage.input_video_duration · number

Duration of the input video in seconds, 0.0 when no video input (wan3.0-video tasks)

usage.output_video_duration · number

Duration of the output video in seconds (wan3.0-video tasks)

usage.ratio · string

Aspect ratio of the generated video, e.g. 16:9 (wan3.0-video tasks)

usage.size · string

Image resolution (T2I tasks)

usage.video_count · integer

Number of generated videos (T2V tasks)

usage.video_duration · number

Duration of generated video in seconds (T2V tasks)

usage.video_ratio · string

Video resolution ratio (T2V tasks)

code · string

Error code for a failed request, reported at the ROOT of the envelope rather than under output (not returned if the request succeeded).

message · string

Detailed information about a failed request, reported at the ROOT of the envelope rather than under output (not returned if the request succeeded). Read this before falling back to output.message.

Examples

Input

{
  "input": {
    "images": [
      "https://example.invalid/red-maple-leaf.png"
    ],
    "prompt": "Make the leaf golden."
  },
  "parameters": {
    "n": 1,
    "size": "768*768"
  }
}

Output

{
  "output": {
    "end_time": "2027-01-01T00:00:12.000Z",
    "results": [
      {
        "actual_prompt": "a single red maple leaf resting on still water, shallow depth of field, soft morning light",
        "orig_prompt": "a single red maple leaf resting on still water",
        "url": "https://example.invalid/wan/wan2.5-t2i-preview/generated-1.png"
      },
      {
        "code": "DataInspectionFailed",
        "message": "This candidate was rejected; the task as a whole succeeded.",
        "orig_prompt": "a single red maple leaf resting on still water"
      }
    ],
    "scheduled_time": "2027-01-01T00:00:01.000Z",
    "submit_time": "2027-01-01T00:00:00.000Z",
    "task_id": "0385dc79-5ff8-4d82-bcb6-7c1a9f2e4d60",
    "task_metrics": {
      "FAILED": 1,
      "SUCCEEDED": 1,
      "TOTAL": 2
    },
    "task_status": "SUCCEEDED"
  },
  "request_id": "7574ee8f-38a3-4b1e-9280-11c33ab46e51",
  "usage": {
    "image_count": 1,
    "size": "1280*1280"
  }
}

Before you ship

The snippets above are the shortest working call. Three things are the same for every model and are documented once on the Comfy Router headers page: send an Idempotency-Key on every paid call and reuse it when you retry, expect the connection to be held up to Router’s 10 minute deadline, and keep X-Comfy-Request-Id from every response. The SDKs do all three for you; the cURL tab does none of them. On failure, X-Comfy-Error-Type names the bucket, and a 422 means the body failed the model’s schema and was never billed.

Headers

Authentication, idempotency, request IDs, error buckets, retry pacing, spend limits.

Quick Start

Typed error handling in Python and TypeScript, reading the 422, walking the catalog.

Limitations

What Router does not do today, and what to use instead.