Images Edits

POST https://wrouter.ai/v1/images/edits

Two request shapes are accepted, routed by model:

• OpenAI shape (multipart/form-data) — dall-e-2, gpt-image-1
• Bailian / Qwen shape (application/json) — qwen-image-edit; the shape natively documented by the underlying gateway

Shape A: OpenAI (multipart/form-data)

Field	Type	Required	Notes
model	string	✓	dall-e-2, gpt-image-1
image	file	✓	PNG with alpha, ≤ 4 MB
mask	file		PNG; transparent pixels = areas to repaint
prompt	string	✓	Description
n	integer
size	string		e.g. 1024x1024
response_format	string		"url" or "b64_json"

curl https://wrouter.ai/v1/images/edits \
  -H "Authorization: Bearer $WROUTER_API_KEY" \
  -F model=gpt-image-1 \
  -F image=@./original.png \
  -F mask=@./mask.png \
  -F prompt="Replace background with a starry sky" \
  -F size=1024x1024

from openai import OpenAI
client = OpenAI(api_key="sk-...", base_url="https://wrouter.ai/v1")
with open("original.png","rb") as img, open("mask.png","rb") as mask:
    out = client.images.edit(
        model="gpt-image-1",
        image=img,
        mask=mask,
        prompt="Tropical beach background",
        size="1024x1024",
    )
print(out.data[0].url)

Shape B: Bailian / Qwen (application/json)

Parameter	Type	Required	Notes
model	string	✓	qwen-image-edit family
input	object	✓	prompt, image (URL or base64), optional mask
parameters	object		size, n, …

curl https://wrouter.ai/v1/images/edits \
  -H "Authorization: Bearer $WROUTER_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "qwen-image-edit",
    "input": {
      "prompt": "Replace background with a starry sky",
      "image": "https://example.com/original.png"
    },
    "parameters": {"size": "1024*1024"}
  }'

Mask rules (OpenAI shape)

• Same dimensions as image
• Transparent pixels = repaint
• Opaque pixels = preserve
• Without mask, the entire image is reinterpreted as reference (model-dependent)