跳转到主要内容
POST
/
image
/
multi-edit
/api/v1/image/multi-edit
curl --request POST \
  --url https://api.venice.ai/api/v1/image/multi-edit \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "prompt": "<string>",
  "images": [
    "<string>"
  ],
  "aspect_ratio": "16:9",
  "modelId": "firered-image-edit",
  "output_format": "png",
  "quality": "high",
  "resolution": "1K",
  "safe_mode": false
}
'
"<string>"
定价： 多图像编辑的价格因模型而异。当前每次编辑的价格请参阅 定价概览。

​
质量等级

部分多图像编辑模型接受可选的 quality 参数（low、medium、high），用于在视觉保真度和成本之间取得平衡。当前由 gpt-image-2-edit 支持；其他模型会忽略该参数。 
{
  "modelId": "gpt-image-2-edit",
  "images": ["iVBORw0KGgo...", "iVBORw0KGgo..."],
  "prompt": "blend the two scenes into one cinematic frame",
  "resolution": "2K",
  "quality": "medium"
}
当您省略 quality 时，模型将使用其默认等级（gpt-image-2-edit 为 high）。各等级价格（1K/2K/4K × low/medium/high）位于 Models 端点中 model_spec.pricing.quality 下，也在定价概览中列出。
质量感知定价是所有 API 和 SDK 调用方的默认行为 —— 无需显式启用的请求头。省略 quality 的请求按模型的默认等级（gpt-image-2-edit 为 high）计费。

授权

​
Authorization
string
header
必填

Bearer authentication header of the form Bearer <token>, where <token> is your auth token.

请求体

Edit an image by compositing up to three layered images with a single prompt. Supports base64-encoded strings and URLs.

​
prompt
string
必填

The text directions to edit or modify the image. Short, descriptive prompts work best (e.g., "remove the tree", "change the sky to sunrise"). Character limit is model specific and is listed in the promptCharacterLimit setting in the model list endpoint.

Required string length: 1 - 32768
​
images
(string<uri> | string)[]
必填

Array of 1 to 3 images used for multi-editing. The first image is treated as the base image, and the remaining images are used as edit layers/masks. Each image can be a base64-encoded string or a URL starting with http:// or https://. Image dimensions must be at least 65536 pixels and must not exceed 33177600 pixels. File size must be less than 25MB.

Required array length: 1 - 3 elements

Image as a base64-encoded string or a URL starting with http:// or https://

​
aspect_ratio
enum<string>

The aspect ratio for the output image. Use 'auto' or omit this parameter to infer the closest supported aspect ratio from the first input image when explicit sizing is required by the model.

可用选项:
auto,
1:1,
3:2,
16:9,
21:9,
9:16,
2:3,
3:4,
4:5
示例:

"16:9"

​
modelId
enum<string>
默认值:firered-image-edit

The model ID to use for multi-edit.

可用选项:
firered-image-edit,
qwen-edit,
qwen-edit-uncensored,
grok-imagine-edit,
grok-imagine-quality-edit,
qwen-image-2-edit,
qwen-image-2-pro-edit,
wan-2-7-pro-edit,
flux-2-max-edit,
gpt-image-2-edit,
gpt-image-1-5-edit,
nano-banana-2-edit,
nano-banana-pro-edit,
seedream-v5-lite-edit,
seedream-v4-edit
Minimum string length: 1
​
output_format
enum<string>

Output format for the edited image. Accepts jpeg, jpg, png, or webp. When omitted, the format is inferred from resolution: PNG for 1K edits and JPEG for 2K/4K edits.

可用选项:
jpeg,
png,
webp
示例:

"png"

​
quality
enum<string>

Output quality for supported models (e.g. GPT Image 2 / GPT Image 2 Edit). Higher values can increase the final request charge. See the model list for supported options.

可用选项:
low,
medium,
high
示例:

"high"

​
resolution
string

Resolution tier for the output image (e.g. "1K", "2K", "4K"). Supported values vary by model - check GET /api/v1/models for model-specific options. Defaults to "1K" when not specified.

Required string length: 1 - 10
示例:

"1K"

​
safe_mode
boolean
默认值:true

Whether to use safe mode. If enabled, this will blur images that are classified as having adult content.

示例:

false

响应

OK

The response is of type file.