Edit/process image (OpenAI SDK)

API Reference

Edit/process image (OpenAI SDK)

Process images using the OpenAI SDK client.images.edit() method. The model parameter determines which AI processing is applied.

Setup

from openai import OpenAI
client = OpenAI(api_key='sk-snap-xxx', base_url='https://api.snapapi.ai/v1')

Supported Models

Removal

Model	Description	`prompt`	`mask`	Credit
`snapapi/remove-bg`	Remove background	not used	-	1
`snapapi/remove-logo`	Remove logo (auto-detect)	not used	optional	5
`snapapi/remove-object`	Remove object by mask	not used	required	3-13
`snapapi/remove-text`	Remove text by mask	not used	required	4
`snapapi/remove-wire`	Remove wires by mask	not used	required	4

Enhance & Restore

Model	Description	`prompt`	Credit
`snapapi/enhance-2x`	Upscale 2x	not used	6
`snapapi/enhance-4x`	Upscale 4x	not used	11
`snapapi/enhance-art-2x`	Upscale art/anime 2x	not used	2
`snapapi/enhance-art-4x`	Upscale art/anime 4x	not used	4
`snapapi/restore`	Restore old photo	not used	1
`snapapi/restore-pro`	Restore (Qwen AI)	not used	13
`snapapi/colorize`	Colorize B&W photo	not used	1
`snapapi/colorize-pro`	Colorize (Qwen AI)	not used	13
`snapapi/light-restore`	Fix lighting issues	not used	21

Generation from Image

Model	Description	`prompt`	Credit
`snapapi/headshot`	Professional headshot	outfit/style description	21
`snapapi/sticker`	Cartoon/sticker	style description	21
`snapapi/qwen-edit`	General AI edit	editing instruction	13
`snapapi/flux-kontext`	FLUX Kontext edit	editing instruction	18

Style-based (prompt = style_id)

Model	Description	`prompt`	Credit
`snapapi/hairstyle`	Transform hairstyle	style_id from list	6
`snapapi/transfer-makeup`	Apply makeup style	style_id from list	1
`snapapi/art`	Transform to AI art	style_id from list	6

Detection (returns JSON, not image)

Model	Description	`prompt`	Credit
`snapapi/detect-objects`	Detect removable objects	not used	1
`snapapi/detect-text`	Detect text regions	not used	1
`snapapi/detect-wires`	Detect wires/cables	not used	1

Utility

Model	Description	`prompt`	`mask`	Credit
`snapapi/retouch-skin`	Skin retouching	not used	-	2
`snapapi/outpaint`	Expand image	not used	required	15
`snapapi/pose-suggest`	Pose suggestion	not used	-	28
`snapapi/qwen-edit-multi`	Multi-image edit	editing instruction	-	26

Examples

Remove background:

result = client.images.edit(
    model='snapapi/remove-bg',
    image=open('photo.png', 'rb'),
    prompt=''
)
print(result.data[0].url)

Enhance 4x:

result = client.images.edit(
    model='snapapi/enhance-4x',
    image=open('photo.png', 'rb'),
    prompt=''
)

Headshot generation:

result = client.images.edit(
    model='snapapi/headshot',
    image=open('face.png', 'rb'),
    prompt='Professional studio headshot with soft lighting'
)

Hairstyle (prompt = style_id):

result = client.images.edit(
    model='snapapi/hairstyle',
    image=open('face.png', 'rb'),
    prompt='ponytail'  # style_id from styles list
)

Pose suggest (extra params via extra_body):

result = client.images.edit(
    model='snapapi/pose-suggest',
    image=open('scene.png', 'rb'),
    prompt='',
    extra_body={'num_models': 3, 'gender': 'female'}
)

Multi-image edit (extra images via extra_body):

result = client.images.edit(
    model='snapapi/qwen-edit-multi',
    image=open('img1.png', 'rb'),
    prompt='Combine the two people in a park',
    extra_body={'input_image_1': 'https://example.com/img2.png'}
)

Detect objects (returns JSON):

result = client.images.edit(
    model='snapapi/detect-objects',
    image=open('photo.png', 'rb'),
    prompt=''
)
print(result.model_dump())  # raw detection JSON

post/v1/images/edits (OpenAI SDK)

Authorization — This endpoint requires an API key. Pass it in the api-key header on every request. See the quickstart

Body

Name	Type	Status	Description
`model`	string	Required	Model to use. See supported models table above.
`image`	file	Required	Input image file.
`prompt`	string	Optional	Depends on model: editing instruction, style_id, or empty string. See table above.
`mask`	file	Optional	Mask image. Required for: `remove-object`, `remove-text`, `remove-wire`, `outpaint`.

Responses

Image processed successfully (or detection JSON for detect-* models)SingleImageResponse

Name	Type	Description
`created`	integer	Unix timestamp of when the response was created
`data`	object[]
`url`	string	URL to download the result image

curl -X POST "https://api.snapapi.ai/v1/images/edits (OpenAI SDK)" \
  -H "api-key: YOUR_API_KEY" \
  -F "model=value" \
  -F "image=@./image.jpg" \
  -F "mask=@./image.png"

Response

{
  "created": 1745827200,
  "data": [
    {
      "url": "https://outputs.snapapi.ai/outputs/abc123.png"
    }
  ]
}