API Reference

SnapAPI - AI Image Processing API

v1.0.0

SnapAPI provides a comprehensive suite of AI-powered image processing endpoints including object removal, image enhancement, background removal, virtual try-on, image editing, and more.

Authentication

Include api-key: YOUR_API_KEY header

Get API Key

Base URL

https://api.snapapi.ai

Credits

Based on your plan

Get your API key

Make your first API call

All endpoints use input_image for image input and return output_image_url — one consistent pattern.

bash

# Option A: api-key header
curl -X POST "https://api.snapapi.ai/v1/images/remove-background" \
  -H "api-key: sk-snap-xxxxx" \
  -F "input_image=@photo.jpg"

# Option B: Bearer token (OpenAI-compatible)
curl -X POST "https://api.snapapi.ai/v1/images/remove-background" \
  -H "Authorization: Bearer sk-snap-xxxxx" \
  -F "input_image=@photo.jpg"

Get your result

Every image endpoint returns OpenAI-compatible format. Your result URL is at data[0].url. No base64 decoding needed.

json

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

More examples

AI Image Editing

bash

# Edit an image with AI prompt
curl -X POST "https://api.snapapi.ai/v1/images/edits" \
  -H "api-key: YOUR_API_KEY" \
  -F "input_image=@photo.jpg" \
  -F "prompt=Transform to anime style" \
  -F "mode=editing"

Text to Image

bash

# Generate image from text
curl -X POST "https://api.snapapi.ai/v1/images/generations" \
  -H "api-key: YOUR_API_KEY" \
  -F "prompt=A professional headshot of a woman in a studio" \
  -F "aspect_ratio=3:4"

Rate Limits

60 requests/minute per API key. Check x-ratelimit-remaining-requests header to track usage. On 429, wait for retry-after seconds.

Image Input

Most endpoints accept images either as a file upload (`input_image`) or as a reference to a previously uploaded image (`session_id` / `image_id`). Mask inputs use `input_mask`.

Need help? SnapAPI Support

Detection

Detect objects, text, and wires in images for use with removal endpoints.

detectObjectsdetectTextdetectWires

post/v1/images/detect-objects

Detect objects in image

Automatically detect removable objects in an image using AI. Returns bounding boxes and masks for detected objects (sorted by confidence score in descending order) and these information can be used in object removal API.

Model: object-detection | Credit: 1

Legacy: /api/object_removal/v5/ai_detection

Parameters

Name	Type	Description
`input_image`	file	The image file (or URL) to detect objects in.
`lang`	string	Language for object class names in response messages (e.g., `en`, `vi`).

Responses

200: Objects detected successfullyDetectionResponse
400: Missing required image inputErrorResponse
429: Rate limit exceeded or insufficient creditsErrorResponse
500: Internal server errorErrorResponse

curl -X POST "https://api.snapapi.ai/v1/images/detect-objects" \
  -H "api-key: YOUR_API_KEY" \
  -F "input_image=@./image.jpg"

Response

{
  "session_id": "sess_abc123",
  "detected_objects": [
    {
      "bbox": [
        120,
        50,
        300,
        400
      ],
      "mask": "base64_encoded_data...",
      "accuracy": 0.95,
      "object_type": "person",
      "object_description": "person"
    }
  ]
}

Schema

session_idstringSession ID for the processed image. Pass as original_session_id to erase endpoint.

detected_objectsobject[]Detected regions sorted by confidence (descending)

bboxnumber[]Bounding box [x1, y1, x2, y2]

maskstringBase64-encoded RGBG mask (same resolution as bbox)

accuracynumberConfidence score (0.0 to 1.0)

object_typestringDetection category

object_descriptionstringDetection class name

post/v1/images/detect-text

Detect text in image

Detect text regions in an image. Returns masks for detected text areas that can be used with the text removal API.

Model: remove-text | Credit: 4

Legacy: /api/object_removal/v5/text_detection

Parameters

Name	Type	Description
`input_image`	file	The image file (or URL) to detect text in.

Responses

200: Text detected successfullySimpleDetectionResponse
400: Missing required image inputErrorResponse
429: Rate limit exceeded or insufficient creditsErrorResponse
500: Internal server errorErrorResponse

curl -X POST "https://api.snapapi.ai/v1/images/detect-text" \
  -H "api-key: YOUR_API_KEY" \
  -F "input_image=@./image.jpg"

Response

{
  "detected": true,
  "mask": "base64_encoded_data..."
}

Schema

detectedbooleanWhether the target was detected

maskstringBase64-encoded RGBG mask. Null if nothing detected. Pass as input_mask to erase endpoint.

post/v1/images/detect-wires

Detect wires/lines in image

Detect power lines, cables, and wires in an image. Returns masks for detected wire regions and these information can be used in wire removal API.

Model: remove-wireline | Credit: 4

Legacy: /api/object_removal/v5/wire_detection

Parameters

Name	Type	Description
`input_image`	file	The image file (or URL) to detect wires in.

Responses

200: Wires detected successfullySimpleDetectionResponse
400: Missing required image inputErrorResponse
429: Rate limit exceeded or insufficient creditsErrorResponse
500: Internal server errorErrorResponse

curl -X POST "https://api.snapapi.ai/v1/images/detect-wires" \
  -H "api-key: YOUR_API_KEY" \
  -F "input_image=@./image.jpg"

Response

{
  "detected": true,
  "mask": "base64_encoded_data..."
}

Schema

detectedbooleanWhether the target was detected

maskstringBase64-encoded RGBG mask. Null if nothing detected. Pass as input_mask to erase endpoint.

Removal & Erase

Remove objects, text, wires, watermarks, and backgrounds from images.

removeObjectsremoveTextremoveWatermarkremoveWiresremoveBackground

post/v1/images/remove-objects

Erase objects from image

Remove objects from an image using mask-based or selected-object erasing. Supports normal erase (GAN), super erase (SD), ultra erase (Qwen).

To get the best results, you should combine this API with the object detection API (i.e., select the detected object to erase).

Models: remove-object-gan, remove-object-sd, remove-object-qwen | Credit: 3 (normal), 4 (super), 13 (ultra).

Legacy: /api/object_removal/v5/erase

Parameters

Name	Type	Description
`input_image`	file	The image file (or URL) to erase objects from.
`original_session_id`	string	Original image session ID for erasing selected objects. This ID comes from the object detection API response.
`input_mask`	file	RGBG mask image file (or URL) for manual erase selection.
`mask_objects`	string	JSON string of selected objects array (subset of `detected_objects` data from object detection API response).
`erase_mode`	string	Erase mode. `normal` for normal erase, `super` for super erase, `ultra` for ultra erase. normalsuperultra (default: `normal`)

Responses

200: Objects erased successfullySingleImageResponse
400: Missing required image inputErrorResponse
429: Rate limit exceeded or insufficient creditsErrorResponse
500: Internal server errorErrorResponse

curl -X POST "https://api.snapapi.ai/v1/images/remove-objects" \
  -H "api-key: YOUR_API_KEY" \
  -F "input_image=@./image.jpg" \
  -F "input_mask=@./image.png" \
  -F "erase_mode=normal"

Response

{
  "created": 1745827200,
  "data": [
    {
      "url": "https://outputs.snapapi.ai/outputs/abc123.png"
    }
  ],
  "session_id": "sess_abc123",
  "image_id": "img_abc123"
}

Schema

createdintegerUnix timestamp of when the response was created

dataobject[]

urlstringURL to download the result image

session_idstringSession ID. Pass as `image_id` in subsequent requests to skip re-uploading.

image_idstringImage ID. Pass in subsequent requests to skip re-uploading.

post/v1/images/remove-text

Erase text from image

Remove detected text from an image using a text mask.

Model: remove-text | Credit: 4

Legacy: /api/object_removal/v5/erase_text

Parameters

Name	Type	Description
`input_image`	file	The image file (or URL) to erase text from.
`input_mask`	file	RGBG mask image (or URL) highlighting text regions to remove. Use the `mask` from text detection API response.
`erase_mode`	string	Erase mode. `normal` for normal erase, `super` for super erase, `ultra` for ultra erase. normalsuperultra (default: `normal`)

Responses

200: Text erased successfullySingleImageResponse
400: Missing required image inputErrorResponse
429: Rate limit exceeded or insufficient creditsErrorResponse
500: Internal server errorErrorResponse

curl -X POST "https://api.snapapi.ai/v1/images/remove-text" \
  -H "api-key: YOUR_API_KEY" \
  -F "input_image=@./image.jpg" \
  -F "input_mask=@./image.png" \
  -F "erase_mode=normal"

Response

{
  "created": 1745827200,
  "data": [
    {
      "url": "https://outputs.snapapi.ai/outputs/abc123.png"
    }
  ],
  "session_id": "sess_abc123",
  "image_id": "img_abc123"
}

Schema

createdintegerUnix timestamp of when the response was created

dataobject[]

urlstringURL to download the result image

session_idstringSession ID. Pass as `image_id` in subsequent requests to skip re-uploading.

image_idstringImage ID. Pass in subsequent requests to skip re-uploading.

post/v1/images/remove-watermark

Erase watermark and logo from image

Remove watermarks and logos from an image. This API supports automatic detection and removal, or manual removal within a selected region.

Model: remove-logo | Credit: 5

Legacy: /api/object_removal/v5/erase_watermark

Parameters

Name	Type	Description
`input_image`	file	The image file (or URL) to remove watermark from.
`input_mask`	file	RGBG mask image (or URL) for manual watermark selection. If not provided, the API will automatically detect and remove the watermark.
`predict_mode`	string	Predict mode for the watermark removal. `3.0` is better for general cases while `2.0` is suitable for emoji cases. 2.03.0 (default: `2.0`)

Responses

200: Watermark erased successfullySingleImageResponse
400: Missing required image inputErrorResponse
429: Rate limit exceeded or insufficient creditsErrorResponse
500: Internal server errorErrorResponse

curl -X POST "https://api.snapapi.ai/v1/images/remove-watermark" \
  -H "api-key: YOUR_API_KEY" \
  -F "input_image=@./image.jpg" \
  -F "input_mask=@./image.png" \
  -F "predict_mode=2.0"

Response

{
  "created": 1745827200,
  "data": [
    {
      "url": "https://outputs.snapapi.ai/outputs/abc123.png"
    }
  ],
  "session_id": "sess_abc123",
  "image_id": "img_abc123"
}

Schema

createdintegerUnix timestamp of when the response was created

dataobject[]

urlstringURL to download the result image

session_idstringSession ID. Pass as `image_id` in subsequent requests to skip re-uploading.

image_idstringImage ID. Pass in subsequent requests to skip re-uploading.

post/v1/images/remove-wires

Erase wires/lines from image

Remove detected wires, power lines, and cables from an image.

Model: remove-wireline | Credit: 4

Legacy: /api/object_removal/v5/erase_wire

Parameters

Name	Type	Description
`input_image`	file	The image file (or URL) to erase wires from.
`input_mask`	file	RGBG mask image (or URL) highlighting wire regions to remove. Use the `mask` from wire detection API response.

Responses

200: Wires erased successfullySingleImageResponse
400: Missing required image inputErrorResponse
429: Rate limit exceeded or insufficient creditsErrorResponse
500: Internal server errorErrorResponse

curl -X POST "https://api.snapapi.ai/v1/images/remove-wires" \
  -H "api-key: YOUR_API_KEY" \
  -F "input_image=@./image.jpg" \
  -F "input_mask=@./image.png"

Response

{
  "created": 1745827200,
  "data": [
    {
      "url": "https://outputs.snapapi.ai/outputs/abc123.png"
    }
  ],
  "session_id": "sess_abc123",
  "image_id": "img_abc123"
}

Schema

createdintegerUnix timestamp of when the response was created

dataobject[]

urlstringURL to download the result image

session_idstringSession ID. Pass as `image_id` in subsequent requests to skip re-uploading.

image_idstringImage ID. Pass in subsequent requests to skip re-uploading.

post/v1/images/remove-background

Remove background from image

Remove the background from an image. Returns a mask of the foreground object.

Model: remove-background | Credit: 1

Legacy: /api/rmbg/v1/

Parameters

Name	Type	Description
`input_image`	file	The image file (or URL) to remove background from.

Responses

200: Background removed successfullyRemoveBackgroundResponse
400: Missing required image inputErrorResponse
429: Rate limit exceeded or insufficient creditsErrorResponse
500: Internal server errorErrorResponse

curl -X POST "https://api.snapapi.ai/v1/images/remove-background" \
  -H "api-key: YOUR_API_KEY" \
  -F "input_image=@./image.jpg"

Response

{
  "output_image_url": "https://outputs.snapapi.ai/outputs/abc123.png",
  "box": [
    10,
    20,
    300,
    400
  ]
}

Schema

output_image_urlstringURL to download the result image (RGBA with transparent background)

boxnumber[]Bounding box [x1, y1, x2, y2] of foreground object

Enhance & Restore

Upscale, enhance, restore, and colorize images.

enhanceProenhanceArtProrestoreImagecolorizeImage

post/v1/images/enhance

Enhance and upscale image (Pro)

Upscale and enhance image resolution using AI. Supports 2x and 4x upscaling with optional face enhancement.

Model: enhance-upscale-image | Credit: 6 (2x), 11 (4x)

Legacy: /api/enhance/v1/pro

Parameters

Name	Type	Description
`input_image`	file	The image file (or URL) to enhance.
`zoom_factor`	string	Upscale factor. `2` for 2x, `4` for 4x resolution. 24
`enhance_faces`	string	Enable face enhancement (`true`/`false`).(default: `true`)
`face_model_ids`	string	Comma-separated face model IDs for face-specific enhancement (`0`, `1` and `2`). Example: `0,1`.(default: `0,1,2`)

Responses

200: Image enhanced successfullySingleImageResponse
400: Missing required image inputErrorResponse
429: Rate limit exceeded or insufficient creditsErrorResponse
500: Internal server errorErrorResponse

curl -X POST "https://api.snapapi.ai/v1/images/enhance" \
  -H "api-key: YOUR_API_KEY" \
  -F "input_image=@./image.jpg" \
  -F "zoom_factor=2" \
  -F "enhance_faces=true" \
  -F "face_model_ids=0,1,2"

Response

{
  "created": 1745827200,
  "data": [
    {
      "url": "https://outputs.snapapi.ai/outputs/abc123.png"
    }
  ],
  "session_id": "sess_abc123",
  "image_id": "img_abc123"
}

Schema

createdintegerUnix timestamp of when the response was created

dataobject[]

urlstringURL to download the result image

session_idstringSession ID. Pass as `image_id` in subsequent requests to skip re-uploading.

image_idstringImage ID. Pass in subsequent requests to skip re-uploading.

post/v1/images/enhance-art

Enhance art/cartoon image (Pro)

Upscale and enhance cartoon, anime, or art-style images with a model optimized for non-photographic content.

Model: enhance-art-image | Credit: 2

Legacy: /api/enhance/v1/cartoon/pro

Parameters

Name	Type	Description
`input_image`	file	The art/cartoon image file (or URL) to enhance.
`zoom_factor`	string	Upscale factor. `2` for 2x, `4` for 4x resolution. 24

Responses

200: Image enhanced successfullySingleImageResponse
400: Missing required image inputErrorResponse
429: Rate limit exceeded or insufficient creditsErrorResponse
500: Internal server errorErrorResponse

curl -X POST "https://api.snapapi.ai/v1/images/enhance-art" \
  -H "api-key: YOUR_API_KEY" \
  -F "input_image=@./image.jpg" \
  -F "zoom_factor=2"

Response

{
  "created": 1745827200,
  "data": [
    {
      "url": "https://outputs.snapapi.ai/outputs/abc123.png"
    }
  ],
  "session_id": "sess_abc123",
  "image_id": "img_abc123"
}

Schema

createdintegerUnix timestamp of when the response was created

dataobject[]

urlstringURL to download the result image

session_idstringSession ID. Pass as `image_id` in subsequent requests to skip re-uploading.

image_idstringImage ID. Pass in subsequent requests to skip re-uploading.

post/v1/images/restore

Restore old images (Normal)

Restore and enhance old images, remove the scratches.

Model: restore-image | Credit: 1

Legacy: /api/restore/v1/

Parameters

Name	Type	Description
`input_image`	file	The old image file (or URL) to restore.

Responses

200: Image restored successfullySingleImageResponse
400: Missing required image inputErrorResponse
429: Rate limit exceeded or insufficient creditsErrorResponse
500: Internal server errorErrorResponse

curl -X POST "https://api.snapapi.ai/v1/images/restore" \
  -H "api-key: YOUR_API_KEY" \
  -F "input_image=@./image.jpg"

Response

{
  "created": 1745827200,
  "data": [
    {
      "url": "https://outputs.snapapi.ai/outputs/abc123.png"
    }
  ],
  "session_id": "sess_abc123",
  "image_id": "img_abc123"
}

Schema

createdintegerUnix timestamp of when the response was created

dataobject[]

urlstringURL to download the result image

session_idstringSession ID. Pass as `image_id` in subsequent requests to skip re-uploading.

image_idstringImage ID. Pass in subsequent requests to skip re-uploading.

post/v1/images/colorize

Colorize old images (Normal)

Colorize old images.

Model: colorize-image | Credit: 1

Legacy: /api/colorize/v1/

Parameters

Name	Type	Description
`input_image`	file	The old image file (or URL) to colorize.

Responses

200: Image colorized successfullySingleImageResponse
400: Missing required image inputErrorResponse
429: Rate limit exceeded or insufficient creditsErrorResponse
500: Internal server errorErrorResponse

curl -X POST "https://api.snapapi.ai/v1/images/colorize" \
  -H "api-key: YOUR_API_KEY" \
  -F "input_image=@./image.jpg"

Response

{
  "created": 1745827200,
  "data": [
    {
      "url": "https://outputs.snapapi.ai/outputs/abc123.png"
    }
  ],
  "session_id": "sess_abc123",
  "image_id": "img_abc123"
}

Schema

createdintegerUnix timestamp of when the response was created

dataobject[]

urlstringURL to download the result image

session_idstringSession ID. Pass as `image_id` in subsequent requests to skip re-uploading.

image_idstringImage ID. Pass in subsequent requests to skip re-uploading.

Generation

Generate images from text prompts or transform photos into AI art.

generateArtgenerateImagegenerateImageQwen

post/v1/images/generate-art

Generate AI art from photo

Transform a photo into AI-generated art using various styles. The style list and preview images are in this link.

The styles can be divided into two categories: v1 (global image transformation) and v2 (avatar-oriented transformation). v1 is preferred for images with multiple subjects, while v2 is designed for a single person.

Model: fairy-ai-art-gen | Credit: 6

Legacy: /api/fairyai/v1/gen_image

Parameters

Name	Type	Description
`input_image`	file	The source photo (or URL) for AI art generation. Provide this or `image_id`.
`image_id`	string	Image ID of a previous AI Art generation request. This is useful for accelerating multiple style transformation requests using the same input image.
`style`	string	Style ID for the art generation.(default: `special_effect_fire`)

Responses

200: Art image generated successfullySingleImageResponse
400: Missing required image inputErrorResponse
429: Rate limit exceeded or insufficient creditsErrorResponse
500: Internal server errorErrorResponse

curl -X POST "https://api.snapapi.ai/v1/images/generate-art" \
  -H "api-key: YOUR_API_KEY" \
  -F "input_image=@./image.jpg" \
  -F "style=special_effect_fire"

Response

{
  "created": 1745827200,
  "data": [
    {
      "url": "https://outputs.snapapi.ai/outputs/abc123.png"
    }
  ],
  "session_id": "sess_abc123",
  "image_id": "img_abc123"
}

Schema

createdintegerUnix timestamp of when the response was created

dataobject[]

urlstringURL to download the result image

session_idstringSession ID. Pass as `image_id` in subsequent requests to skip re-uploading.

image_idstringImage ID. Pass in subsequent requests to skip re-uploading.

post/v1/images/generations

AI image generation using Z-Image model.

Generate images using natural language prompts powered by the Z-Image model. This API is particularly strong at generating realistic human images.

Models: generate-of-z-image| Credit: 9

Legacy: /api/text-to-image/v1

Parameters

Name	Type	Description
`prompt`	string	Natural language description of the desired edit(default: )
`aspect_ratio`	string	Aspect ratio (height-width) of the generated image 1:13:22:34:33:416:99:16 (default: `1:1`)

Responses

200: Image generated successfullySingleImageResponse
400: Missing required image inputErrorResponse
429: Rate limit exceeded or insufficient creditsErrorResponse
500: Internal server errorErrorResponse

curl -X POST "https://api.snapapi.ai/v1/images/generations" \
  -H "api-key: YOUR_API_KEY" \
  -F "prompt=" \
  -F "aspect_ratio=1:1"

Response

{
  "created": 1745827200,
  "data": [
    {
      "url": "https://outputs.snapapi.ai/outputs/abc123.png"
    }
  ],
  "session_id": "sess_abc123",
  "image_id": "img_abc123"
}

Schema

createdintegerUnix timestamp of when the response was created

dataobject[]

urlstringURL to download the result image

session_idstringSession ID. Pass as `image_id` in subsequent requests to skip re-uploading.

image_idstringImage ID. Pass in subsequent requests to skip re-uploading.

post/v1/images/generations-qwen

AI image generation using Qwen-Image model.

Generate an image using natural language prompts powered by the Qwen-Image model. This API is suitable for general cases, including poster, illustration, anime images.

Models: generate-of-qwen-image| Credit: 9

Legacy: /api/text-to-image/v2

Parameters

Name	Type	Description
`prompt`	string	Natural language description of the desired edit(default: )
`aspect_ratio`	string	Aspect ratio (height-width) of the generated image 1:13:22:34:33:416:99:16 (default: `1:1`)

Responses

200: Image generated successfullySingleImageResponse
400: Missing required image inputErrorResponse
429: Rate limit exceeded or insufficient creditsErrorResponse
500: Internal server errorErrorResponse

curl -X POST "https://api.snapapi.ai/v1/images/generations-qwen" \
  -H "api-key: YOUR_API_KEY" \
  -F "prompt=" \
  -F "aspect_ratio=1:1"

Response

{
  "created": 1745827200,
  "data": [
    {
      "url": "https://outputs.snapapi.ai/outputs/abc123.png"
    }
  ],
  "session_id": "sess_abc123",
  "image_id": "img_abc123"
}

Schema

createdintegerUnix timestamp of when the response was created

dataobject[]

urlstringURL to download the result image

session_idstringSession ID. Pass as `image_id` in subsequent requests to skip re-uploading.

image_idstringImage ID. Pass in subsequent requests to skip re-uploading.

Editing

Edit images using AI prompts with various models and modes.

editImageeditImageMultieditFlux

post/v1/images/edits

Edit image with AI prompt

Edit an image using natural language prompts powered by the Qwen-Image-Edit model. Supports multiple editing modes:

Mode	Description	Example
editing	General editing prompt	Transform the image to anime style
headshot	Professional portrait image generation for the given person image. You need to add [FACE CROP] to the prompt	[FACE CROP] Professional headshot image of the reference subject in a studio
cartoon	Create a cartoon/sticker-style image	Turn the characters in the image into Apple iOS-style 3D avatars
light_restore	Fix the lighting of an overexposed image. Can be run with an empty prompt	(empty prompt)
restore	Restore old or degraded images (e.g., scratches, noise). Can be run with an empty prompt	(empty prompt)
colorize	Add color to old black-and-white images. Can be run with an empty prompt	(empty prompt)
inpaint	Edit a specific region in the input image. Need to provide `input_mask` in the request data	Add a ball

Models: general-edit-of-qwen-image-edit, headshot-generation-of-qwen-image-edit, cartoon-sticker-generation-of-qwen-image-edit, light-restoration-of-qwen-image-edit, restore-of-qwen-image-edit, colorize-of-qwen-image-edit, inpaint-of-qwen-image-edit | Credit: 13 (editing/restore/colorize/inpaint), 21 (headshot/cartoon/light_restore)

Legacy: /api/image-edit/v2

Parameters

Name	Type	Description
`input_image`	file	The image file (or URL) to edit.
`prompt`	string	Natural language description of the desired edit
`mode`	string	Editing mode to apply. editingheadshotcartoonlight_restoreinpaint (default: `editing`)
`strength`	number	Strength of the editing effect (0.0 to 1.0).(default: `1`)
`guidance_scale`	number	Guidance scale for the model. Higher values produce results closer to the prompt.(default: `1`)
`inference_steps`	integer	Number of inference steps. More steps = higher quality but slower.(default: `4`)

Responses

200: Image edited successfullySingleImageResponse
400: Missing required image inputErrorResponse
429: Rate limit exceeded or insufficient creditsErrorResponse
500: Internal server errorErrorResponse

curl -X POST "https://api.snapapi.ai/v1/images/edits" \
  -H "api-key: YOUR_API_KEY" \
  -F "input_image=@./image.jpg" \
  -F "mode=editing" \
  -F "strength=1" \
  -F "guidance_scale=1" \
  -F "inference_steps=4"

Response

{
  "created": 1745827200,
  "data": [
    {
      "url": "https://outputs.snapapi.ai/outputs/abc123.png"
    }
  ],
  "session_id": "sess_abc123",
  "image_id": "img_abc123"
}

Schema

createdintegerUnix timestamp of when the response was created

dataobject[]

urlstringURL to download the result image

session_idstringSession ID. Pass as `image_id` in subsequent requests to skip re-uploading.

image_idstringImage ID. Pass in subsequent requests to skip re-uploading.

post/v1/images/edits/multi

Edit based on multiple input images with AI prompt

Edit based on multiple images (up to 3) using natural language prompts powered by the Qwen-Image-Edit model. Supports multiple editing modes:

Mode	Description	Example
editing	General editing prompt	Put the man and the dog in a stadium

Models: general-edit-multi-of-qwen-image-edit | Credit: 26

Legacy: /api/image-edit/v2/multi

Parameters

Name	Type	Description
`input_image_0`	file	The first image file to edit.
`input_image_1`	file	The second image file to edit (optional).
`input_image_2`	file	The third image file to edit (optional).
`prompt`	string	Natural language description of the desired edit
`mode`	string	Editing mode to apply. Only `editing` is supported at this time. editing (default: `editing`)
`strength`	number	Strength of the editing effect (0.0 to 1.0).(default: `1`)
`guidance_scale`	number	Guidance scale for the model. Higher values produce results closer to the prompt.(default: `1`)
`inference_steps`	integer	Number of inference steps. More steps = higher quality but slower.(default: `4`)

Responses

200: Image edited successfullySingleImageResponse
400: Missing required image inputErrorResponse
429: Rate limit exceeded or insufficient creditsErrorResponse
500: Internal server errorErrorResponse

curl -X POST "https://api.snapapi.ai/v1/images/edits/multi" \
  -H "api-key: YOUR_API_KEY" \
  -F "input_image_0=@./image.jpg" \
  -F "input_image_1=@./image.jpg" \
  -F "input_image_2=@./image.jpg" \
  -F "mode=editing" \
  -F "strength=1" \
  -F "guidance_scale=1" \
  -F "inference_steps=4"

Response

{
  "created": 1745827200,
  "data": [
    {
      "url": "https://outputs.snapapi.ai/outputs/abc123.png"
    }
  ],
  "session_id": "sess_abc123",
  "image_id": "img_abc123"
}

Schema

createdintegerUnix timestamp of when the response was created

dataobject[]

urlstringURL to download the result image

session_idstringSession ID. Pass as `image_id` in subsequent requests to skip re-uploading.

image_idstringImage ID. Pass in subsequent requests to skip re-uploading.

post/v1/images/edit-flux

Create image editing task (FLUX Kontext)

Create an async image editing task using the FLUX Kontext model. Currently, only the editing mode is supported for general image editing use cases.

Models: general-image-editing-of-flux-kontext | Credit: 18

Legacy: /api/image-editing/v1/tasks

Parameters

Name	Type	Description
`input_image`	file	The image file (or URL) to edit. Provide this or `image_id`.
`image_id`	string	Session ID of a previously uploaded image.
`prompt`*	string	Text prompt describing the desired edit.
`mode`	string	LoRA mode for the editing task. Only `editing` is supported at this time.(default: `editing`)
`strength`	number	LoRA strength (0.0 to 1.0).(default: `1`)
`guidance_scale`	number	Guidance scale for the model.(default: `3.5`)
`inference_steps`	integer	Number of inference steps.(default: `8`)
`run_fast`	string	Enable fast mode for quicker but potentially lower quality results. truefalse (default: `false`)

Responses

200: Image editing task createdTaskCreatedResponse
400: Missing required fields or invalid modeErrorResponse
429: Rate limit exceeded or insufficient creditsErrorResponse
500: Internal server errorErrorResponse

curl -X POST "https://api.snapapi.ai/v1/images/edit-flux" \
  -H "api-key: YOUR_API_KEY" \
  -F "input_image=@./image.jpg" \
  -F "prompt=value" \
  -F "mode=editing" \
  -F "strength=1" \
  -F "guidance_scale=3.5" \
  -F "inference_steps=8" \
  -F "run_fast=true"

Response

{
  "task_id": "task_abc123",
  "status": "CREATED",
  "created_at": 1745827200
}

Schema

task_idstringUnique task identifier. Poll GET /v1/tasks/{task_id} for results.

statusstring

created_atintegerUnix timestamp of task creation

Beauty & Style

AI-powered skin retouching, makeup transfer, and hairstyle transformation.

retouchSkintransferMakeuphairstyle

post/v1/images/retouch-skin

Skin retouching

Apply natural skin retouching and enhancement to all faces in the image.

Model: skin-natural-retouch | Credit: 2

Legacy: /api/skin_beauty/v1

Parameters

Name	Type	Description
`input_image`	file	The image file (or URL) for beauty enhancement.

Responses

200: Skin beauty applied successfullySingleImageResponse
400: Missing required image inputErrorResponse
429: Rate limit exceeded or insufficient creditsErrorResponse
500: Internal server errorErrorResponse

curl -X POST "https://api.snapapi.ai/v1/images/retouch-skin" \
  -H "api-key: YOUR_API_KEY" \
  -F "input_image=@./image.jpg"

Response

{
  "created": 1745827200,
  "data": [
    {
      "url": "https://outputs.snapapi.ai/outputs/abc123.png"
    }
  ],
  "session_id": "sess_abc123",
  "image_id": "img_abc123"
}

Schema

createdintegerUnix timestamp of when the response was created

dataobject[]

urlstringURL to download the result image

session_idstringSession ID. Pass as `image_id` in subsequent requests to skip re-uploading.

image_idstringImage ID. Pass in subsequent requests to skip re-uploading.

post/v1/images/transfer-makeup

Makeup transfer

Apply makeup styles to all faces in the image. The style list and preview images are in this link.

Model: makeup-transfer | Credit: 1

Legacy: /api/makeup/v1/transfer

Parameters

Name	Type	Description
`input_image`	file	The image file (or URL) for makeup transfer.
`style`	string	Style ID to apply the makeup.(default: `camdao`)

Responses

200: Makeup applied successfullySingleImageResponse
400: Missing required image inputErrorResponse
429: Rate limit exceeded or insufficient creditsErrorResponse
500: Internal server errorErrorResponse

curl -X POST "https://api.snapapi.ai/v1/images/transfer-makeup" \
  -H "api-key: YOUR_API_KEY" \
  -F "input_image=@./image.jpg" \
  -F "style=camdao"

Response

{
  "created": 1745827200,
  "data": [
    {
      "url": "https://outputs.snapapi.ai/outputs/abc123.png"
    }
  ],
  "session_id": "sess_abc123",
  "image_id": "img_abc123"
}

Schema

createdintegerUnix timestamp of when the response was created

dataobject[]

urlstringURL to download the result image

session_idstringSession ID. Pass as `image_id` in subsequent requests to skip re-uploading.

image_idstringImage ID. Pass in subsequent requests to skip re-uploading.

post/v1/images/hairstyle

Generate hairstyle transformation

Transform the hairstyle in a face photo with multiple reference hairstyles in this link.

Model: hairstyle-generation-lora-of-sd-1-5 | Credit: 6

Legacy: /api/hairstyler/v1

Parameters

Name	Type	Description
`input_image`	file	The face image file for hairstyle transformation. Provide this or `image_id`.
`image_id`	string	Session ID of a previously uploaded image. This is useful for accelerating multiple hairstyle transformation requests using the same face image.
`style`	string	Hairstyle ID to apply.

Responses

200: Hairstyle generated successfullySingleImageResponse
400: Missing required image inputErrorResponse
429: Rate limit exceeded or insufficient creditsErrorResponse
500: Internal server errorErrorResponse

curl -X POST "https://api.snapapi.ai/v1/images/hairstyle" \
  -H "api-key: YOUR_API_KEY" \
  -F "input_image=@./image.jpg"

Response

{
  "created": 1745827200,
  "data": [
    {
      "url": "https://outputs.snapapi.ai/outputs/abc123.png"
    }
  ],
  "session_id": "sess_abc123",
  "image_id": "img_abc123"
}

Schema

createdintegerUnix timestamp of when the response was created

dataobject[]

urlstringURL to download the result image

session_idstringSession ID. Pass as `image_id` in subsequent requests to skip re-uploading.

image_idstringImage ID. Pass in subsequent requests to skip re-uploading.

Virtual Try-On

Virtual clothing try-on using AI.

createTryOnTask

post/v1/images/try-on

Create virtual try-on task

Create an asynchronous virtual try-on task. Upload a model (person) image and one or two clothing images to generate the try-on result.

This API supports either:

One clothing item of type upper, lower, or full
Two clothing items simultaneously (upper and lower)

Poll the GET endpoint with the returned task_id to retrieve results.

Model: virtual-tryon-clothes | Credit: 17

Legacy: /api/tryon/v2/tasks

Parameters

Name	Type	Description
`model_image`	file	Person/model image file (or URL). Provide this or `model_image_id`.
`model_image_id`	string	Session ID of a previously uploaded model image. This is useful for accelerating multiple try-on requests using the same model image.
`cloth_image`	file	Clothing image file (or URL). Provide this or `cloth_image_id`.
`cloth_image_id`	string	Session ID of a previously uploaded clothing image. This is useful for accelerating multiple try-on requests using the same clothing image.
`lower_cloth_image`	file	Lower body clothing image file (or URL, if needed). Provide this or `lower_cloth_image_id`.
`lower_cloth_image_id`	string	Session ID of a previously uploaded lower clothing image (if needed). This is useful for accelerating multiple try-on requests using the same clothing image.
`cloth_type`	string	Type of clothing (e.g., `upper`, `lower`, `full`). If using both `cloth_image` and `lower_cloth_image`, this value must be set to `full`.

Responses

200: Try-on task created successfullyTaskCreatedResponse
400: Missing required fieldsErrorResponse
429: Rate limit exceeded or insufficient creditsErrorResponse
500: Internal server errorErrorResponse

curl -X POST "https://api.snapapi.ai/v1/images/try-on" \
  -H "api-key: YOUR_API_KEY" \
  -F "model_image=@./image.jpg" \
  -F "cloth_image=@./image.jpg" \
  -F "lower_cloth_image=@./image.jpg"

Response

{
  "task_id": "task_abc123",
  "status": "CREATED",
  "created_at": 1745827200
}

Schema

task_idstringUnique task identifier. Poll GET /v1/tasks/{task_id} for results.

statusstring

created_atintegerUnix timestamp of task creation

Utility

Utility endpoints including outpaint, pose suggestion, and health check.

outpaintposeSuggesthealthCheck

post/v1/images/outpaint

Expand image (outpaint)

Expand an image beyond its original boundaries. Requires both the source image and a mask indicating the expansion area.

Model: expand-image-outpaint | Credit: 15

Legacy: /api/outpaint/v1/

Parameters

Name	Type	Description
`input_image`	file	The source image file (or URL) containing the original content placed on a white canvas, where white pixels indicate the expansion area.
`input_mask`	file	A grayscale mask image where white pixels indicate the expansion area and black pixels indicate the original content.

Responses

200: Image outpainted successfullySingleImageResponse
400: Missing required image inputErrorResponse
429: Rate limit exceeded or insufficient creditsErrorResponse
500: Internal server errorErrorResponse

curl -X POST "https://api.snapapi.ai/v1/images/outpaint" \
  -H "api-key: YOUR_API_KEY" \
  -F "input_image=@./image.jpg" \
  -F "input_mask=@./image.png"

Response

{
  "created": 1745827200,
  "data": [
    {
      "url": "https://outputs.snapapi.ai/outputs/abc123.png"
    }
  ],
  "session_id": "sess_abc123",
  "image_id": "img_abc123"
}

Schema

createdintegerUnix timestamp of when the response was created

dataobject[]

urlstringURL to download the result image

session_idstringSession ID. Pass as `image_id` in subsequent requests to skip re-uploading.

image_idstringImage ID. Pass in subsequent requests to skip re-uploading.

post/v1/images/pose-suggest

Get pose suggestions

Generate AI pose suggestions for a reference scene image.

Model: pose-suggestion-lora-of-qwen-image-edit-2509 | Credit: 28

Legacy: /api/recommendation/v1/pose

Parameters

Name	Type	Description
`input_image`	file	The reference image file.
`num_models`	integer	Number of models (persons) for pose recommendations.(default: `1`)
`gender`	string	Gender for pose recommendations. femalemale (default: `female`)

Responses

200: Pose suggestions generated successfullySingleImageResponse
400: Missing required image inputErrorResponse
429: Rate limit exceeded or insufficient creditsErrorResponse
500: Internal server errorErrorResponse

curl -X POST "https://api.snapapi.ai/v1/images/pose-suggest" \
  -H "api-key: YOUR_API_KEY" \
  -F "input_image=@./image.jpg" \
  -F "num_models=1" \
  -F "gender=female"

Response

{
  "created": 1745827200,
  "data": [
    {
      "url": "https://outputs.snapapi.ai/outputs/abc123.png"
    }
  ],
  "session_id": "sess_abc123",
  "image_id": "img_abc123"
}

Schema

createdintegerUnix timestamp of when the response was created

dataobject[]

urlstringURL to download the result image

session_idstringSession ID. Pass as `image_id` in subsequent requests to skip re-uploading.

image_idstringImage ID. Pass in subsequent requests to skip re-uploading.

get/v1/health

Health check

Returns the health status of all SnapAPI services. No authentication required.

Responses

200: All services healthy

curl -X GET "https://api.snapapi.ai/v1/health" \
  -H "api-key: YOUR_API_KEY"

Response

{
  "status": "healthy",
  "checks": {
    "worker": "ok",
    "upstream": "ok",
    "backend": "ok",
    "kv": "ok"
  },
  "timestamp": "2026-04-28T10:00:00Z"
}

Schema

statusstring

checksobject

workerstring

upstreamstring

backendstring

kvstring

timestampstring

Tasks

Poll and manage async tasks created by Virtual Try-On and FLUX Kontext editing endpoints.

getTaskStatuscancelTask

get/v1/tasks/{task_id}

Get task status

Retrieve the status and result of an async task (Virtual Try-On or FLUX Kontext editing). Poll this endpoint after creating a task. Auto-fails tasks with CREATED status older than 120 seconds.

Parameters

Name	Type	Description
`task_id`*	string	The task ID returned from the create task endpoint.

Responses

200: Task status retrievedTaskStatusResponse
404: Task not foundErrorResponse

curl -X GET "https://api.snapapi.ai/v1/tasks/YOUR_TASK_ID" \
  -H "api-key: YOUR_API_KEY"

Response

{
  "task_id": "task_abc123",
  "status": "COMPLETED",
  "progress": 75,
  "output_image_url": "https://outputs.snapapi.ai/outputs/result.png",
  "error_msg": "string",
  "created_at": 0,
  "started_at": 0,
  "completed_at": 0
}

Schema

task_idstring

statusstring

progressintegerProgress 0-100 (when PROCESSING)

output_image_urlstringResult image URL (when COMPLETED)

error_msgstringError message (when FAILED)

created_atinteger

started_atinteger

completed_atinteger

delete/v1/tasks/{task_id}

Cancel task

Cancel a pending async task (Virtual Try-On or FLUX Kontext editing).

Parameters

Name	Type	Description
`task_id`*	string	The task ID to cancel.

Responses

200: Task cancelled

curl -X DELETE "https://api.snapapi.ai/v1/tasks/YOUR_TASK_ID" \
  -H "api-key: YOUR_API_KEY"

Response

{
  "message": "task abc123 cancelled"
}

Schema

messagestring