Generate entirely unique AI fashion models or upload custom reference images to use in try-ons.

Model Create

Model Create enables you to generate realistic fashion models with simple, intuitive prompts or reference images, or to upload existing model images to your workspace.

Model Name: model-create

Lifecycle: stable

Processing Time: ~10s–30s (increases when using advanced references like face/pose)

Output Format: Auto (JPEG or PNG)

Delivery Methods: URL

Credits: 1 to 5 per model (1k: 1, 2k: 3, 4k: 5)

Request

Generate or upload fashion models by submitting your configuration to the universal /v1/run endpoint:

POSThttps://fititon.app/api/v1/run

Request Examples

curl -X POST https://fititon.app/api/v1/run \
     -H "Content-Type: application/json" \
     -H "Authorization: Bearer YOUR_API_KEY" \
     -d '{
           "model_name": "model-create",
           "inputs": {
             "type": "generate",
             "name": "Summer Collection Model",
             "prompt": "Full body shot, woman wearing a white t-shirt and dark blue biker shorts, studio lighting",
             "gender": "female",
             "quality": "2k",
             "sampleCount": 1
           }
         }'

Response

Because model generation can take up to 30 seconds, the /v1/run endpoint returns a 201 Created status immediately with an asynchronous job id. You must poll the Status API using GET /v1/status/{id} to retrieve the final results. Alternatively, you can configure webhooks to receive notifications automatically. Please note that our webhook dispatcher enforces a strict 15-second timeout. Your server must acknowledge the webhook with a 2xx status code within 15 seconds, or we will consider it failed and retry it later.

{
  "id": "123a87r9-4129-4bb3-be18-9c9fb5bd7fc1",
  "status": "starting",
  "created_at": "2026-07-18T17:30:00.000Z",
  "error": null,
  "output": null
}

Once the status changes to "succeeded", the output field will contain an array with the URL pointing to your generated model:

{
  "id": "123a87r9-4129-4bb3-be18-9c9fb5bd7fc1",
  "status": "succeeded",
  "created_at": "2026-07-18T17:30:00.000Z",
  "error": null,
  "output": [
    "https://cdn.fititon.app/users/123/results/model-123.png"
  ]
}

Request Parameters

Required Parameters

typeRequired'generate' | 'upload'

Specifies whether to generate a new model via AI (generate) or upload an existing image (upload).

nameRequiredstring

A descriptive name for your model. This is used to identify the model in your studio.

promptstring

Required for generation Text description of the model to generate. Describes the desired fashion model, clothing, pose, and background scene.

Pro Tip: If you are using a reference like faceModelId, the AI will prioritize the reference's identity but will use your prompt to determine the setting, outfit, and body characteristics. Optional only if a faceModelId is provided.

imageBase64string

Required for upload Base64-encoded image string. Required when type is set to upload.

Formatting Requirement Your Base64 string must include the standard data URI prefix (e.g., data:image/jpeg;base64,... or data:image/png;base64,...). The upload will fail if only the raw encoded string is provided.

Optional Generation Parameters

These parameters apply when type is set to generate.

genderstring

The model's gender. Aids the generation model in determining physical attributes.

bodyTypestring

The model's body type (e.g., "athletic", "plus size", "slim").

clothingstring

Specific clothing details to enforce during generation, augmenting the main prompt.

posestring

Textual description of the pose (e.g., "standing hands in pockets", "walking").

faceModelIdstring

UUID of an existing face model to lock in a specific identity across generations. The output will adapt to the target prompt while preserving the reference face.

Note: Supplying a face reference adds several seconds to the total processing time as the AI aligns the identity.

poseModelIdstring

UUID of an existing pose model. Forces the generation to strictly match the reference for body pose and posture.

Note: Supplying a pose reference increases the processing time.

poseImageBase64string

Base64-encoded image string to use as an ad-hoc pose reference instead of providing a saved poseModelId.

quality'1k' | '2k' | '4k'

Output resolution tier. 1k produces ~1 megapixel output, 2k ~4 megapixels, and 4k ~16 megapixels. Higher resolutions consume more credits. Default: 1k

aspectRatiostring

Defines the width-to-height ratio of the generated image.

Supported values: '1:1', '3:4', '4:3', '9:16', '16:9', '2:3', '3:2', '4:5', '5:4', '21:9'.

Default: 3:4

sampleCountinteger

Number of image variations to generate per request. Additional images consume more credits and increase processing time linearly. Default: 1

Runtime Errors

Runtime errors for this feature use the shared set documented in Error Handling. If an error occurs during processing, the status will update to "error" and the error field will contain the specific failure reason.

Credit Cost

For Developer API requests, the cost depends on the chosen output quality tier.

Quality Tier	Credits per Model
1k (Default)	1
2k	3
4k	5