Wan2.7 Image | EmpirioLabs AI Docs

Alibaba Cloud · Image Generation

POST /v1/images/generations

Image generation and editing companion model: text-to-image, bounding-box edits, and cohesive image sets, with up to 4K output on Pro.

At a glance

Field	Value
Model id	`wan2-7-image`
Input modalities	Text, Image
Output modalities	Image
Context window	-
Weight precision	-
Region	Singapore
Features	image_editing, 4k
Native inference	No
New	No
Supported endpoints	`POST /v1/images/generations`

Pricing

Charge	Spec	Rate
Standard	per image	$0.030
Pro	per image	$0.075

Example request

$ curl https://api.empiriolabs.ai/v1/images/generations \
>   -H 'Authorization: Bearer $EMPIRIOLABS_API_KEY' \
>   -H 'Content-Type: application/json' \
>   -d '{"model": "wan2-7-image", "prompt": "a misty forest at dawn", "size": "1024x1024"}'

Parameters

Parameter	Type	Required	Default	Description
`prompt`	string	yes	-	Text prompt. For Image Set mode, describe each image in sequence (e.g. ‘First image: … Second image: …’).
`model_variant`	enum	no	`"wan2.7-image-pro"`	Pro: 4K + Thinking Mode + higher quality. Standard: faster, cheaper. · Allowed: `wan2.7-image-pro`, `wan2.7-image`
`aspect_ratio`	enum	no	-	Optional. If omitted, text-to-image defaults to 16:9; image-edit matches the last input image’s native ratio. · Allowed: `16:9`, `1:1`, `9:16`, `3:2`, `2:3`, `4:3`, `3:4`, `5:4`, `4:5`
`resolution`	enum	no	`"2K"`	Quality tier. 1K ~1MP, 2K ~4MP, 4K ~16MP (Pro text-to-image only). · Allowed: `1K`, `2K`, `4K`
`use_custom_size`	boolean	no	false	Override aspect_ratio + resolution with explicit pixel dimensions.
`custom_width`	number	no	-	Pixels (768–2048; Pro text-to-image up to 4096). Only when use_custom_size=true. Aspect must be in [1:8, 8:1]. · Range: 768 – 4096
`custom_height`	number	no	-	Pixels (768–2048; Pro text-to-image up to 4096). Only when use_custom_size=true. · Range: 768 – 4096
`enable_sequential`	boolean	no	false	Image Set Mode. Generate up to 12 cohesive related images. Disables thinking_mode and color_palette.
`num_images`	number	no	`1`	Standard mode only. Number of images to generate. Each is billed separately. · Range: 1 – 4
`num_images_set`	number	no	`1`	Image Set Mode only. Maximum images in the set. The model decides the actual count; you’re billed only for images generated. · Range: 1 – 12
`thinking_mode`	boolean	no	true	Pro only. Enhanced inference for higher quality at the cost of latency. Auto-ignored when input images are attached or sequential mode is on.
`image`	array	no	-	Up to 9 input image URLs for editing. Triggers image-edit mode automatically.
`color_palette`	string	no	-	Optional JSON array of 3–10 {hex, ratio} objects (ratios sum to 100%). Standard mode only. Example: [{“hex”:“#C2D1E6”,“ratio”:“60.00%”},{“hex”:“#636574”,“ratio”:“40.00%”}]
`bbox_list`	string	no	-	Interactive editing boxes as JSON. Use [] for no box, or [[x1,y1,x2,y2]] for one box.
`seed`	number	no	-	Reproducibility seed. Leave empty for random. · Range: 0 – 2147483647
`watermark`	boolean	no	false	Forwarded to DashScope as parameters.watermark. When true, the upstream Wan service stamps a small Alibaba/Wan provenance watermark on each generated image. EmpirioLabs does not add any watermark of its own.
`response_format`	enum	no	`"url"`	url (default) returns a signed URL; b64_json returns inline base64. · Allowed: `url`, `b64_json`

Notes

Upload 1-9 images for editing mode; without images, operates as text-to-image.

Constraints

Prompt max 5,000 characters
Generation: 10-60s depending on resolution, image count, model variant, and thinking mode

Image inputs (editing mode)

JPEG, PNG, BMP, or WEBP
240-8000 px per side
Aspect ratio 1:8 to 8:1
Max 20 MB per image

Capabilities

Style transfer, object placement, scene blending, background replacement, multi-reference composition, interactive editing with bounding boxes
Pro variant supports up to 4K output

Machine-readable schema: GET https://api.empiriolabs.ai/v1/models/wan2-7-image.