Advanced retouching capabilities - SeedEdit 3.0 i2i

We test the SeedEdit 3.0's photo retouching capabilities.

 1import requests
 2import time
 3import json
 4
 5api_key = "YOUR_API_KEY"
 6url = "https://api.segmind.com/workflows/68a98710c49c91c2edbbbfa0-v1"
 7
 8data = {
 9    "image_ukbxw": "https://segmind-inference-inputs.s3.amazonaws.com/73c6b9f0-2415-4533-99ef-8cfd14db285d-example-retouch-min.png",
10    "Instructions": "Retouch this photoshoot image of a model. Apply the following edits:\n\n1. Skin Retouching: medium, preserve natural skin texture.\n\n2. Lighting & Exposure: brightness +1, contrast +1, shadow recovery low.\n\n3. Color Balance & Tone: skin tone warm, saturation medium.\n\n4. Facial & Body Refinement: subtle adjustments, focus on eyes and hair.\n\n5. Background Cleanup: remove distractions yes, background blur soft.\n\nKeep the result natural and high quality, avoiding over-editing.\n",
11    "Upscale_x": "2x"
12}
13
14def poll_for_result(poll_url):
15    """Poll the API until the generation is complete"""
16    while True:
17        response = requests.get(
18            poll_url,
19            headers={'Authorization': f'Bearer {api_key}'}
20        )
21        result = response.json()
22
23        if result['status'] == 'COMPLETED':
24            # Parse the output (it's a JSON string)
25            outputs = json.loads(result['output'])
26            return outputs
27        elif result['status'] == 'FAILED':
28            raise Exception(result.get('error', 'Generation failed'))
29
30        # Wait 7 seconds before polling again
31        time.sleep(7)
32
33# Make the initial request
34response = requests.post(
35    url,
36    json=data,
37    headers={
38        'Authorization': f'Bearer {api_key}',
39        'Content-Type': 'application/json'
40    }
41)
42
43if response.status_code == 200:
44    result = response.json()
45    print('Request queued:', result)
46
47    # Start polling for results
48    outputs = poll_for_result(result['poll_url'])
49    print('Generation complete:', outputs)
50else:
51    print(f"Error: {response.status_code}")
52    print(response.text)

API Endpoint

POSThttps://api.segmind.com/workflows/68a98710c49c91c2edbbbfa0-v1

Parameters

image_ukbxwoptional
string (uri)

Image Ukbxw

Default: "https://segmind-inference-inputs.s3.amazonaws.com/73c6b9f0-2415-4533-99ef-8cfd14db285d-example-retouch-min.png"
Instructionsoptional
string

Instructions

Default: "Retouch this photoshoot image of a model. Apply the following edits:\n\n1. Skin Retouching: medium, preserve natural skin texture.\n\n2. Lighting & Exposure: brightness +1, contrast +1, shadow recovery low.\n\n3. Color Balance & Tone: skin tone warm, saturation medium.\n\n4. Facial & Body Refinement: subtle adjustments, focus on eyes and hair.\n\n5. Background Cleanup: remove distractions yes, background blur soft.\n\nKeep the result natural and high quality, avoiding over-editing.\n"
Upscale_xoptional
string

Upscale X

Default: "2x"

Response Format

Returns: Polling-based asynchronous response

Initial request returns a poll_url. Poll every 7 seconds until status is COMPLETED.

Common Error Codes

The API returns standard HTTP status codes. Detailed error messages are provided in the response body.

400

Bad Request

Invalid parameters or request format

401

Unauthorized

Missing or invalid API key

403

Forbidden

Insufficient permissions

404

Not Found

Workflow not found

406

Insufficient Credits

Not enough credits to process request

429

Rate Limited

Too many requests

500

Server Error

Internal server error

502

Bad Gateway

Service temporarily unavailable

504

Timeout

Request timed out