bria/expand-image

OfficialView on Replicate →

Bria Expand expands images beyond their borders in high quality. Resizing the image by generating new pixels to expand to the desired aspect ratio. Trained exclusively on licensed data for safe and risk-free commercial use

Capabilities

Reference ImagesNegative PromptSeed

Cost

Community model (estimated from hardware time)

Input Parameters

Name	Type	Description	Default	Constraints
`aspect_ratio`	string	Aspect ratio for expansion. Either aspect_ratio or canvas_size with original_image_size/location must be provided. Can be a predefined string like '1:1', '16:9' etc. or a custom float between 0.5 and 3.0	`"1:1"`	1:12:33:23:44:34:55:49:1616:9
`canvas_size`	array	Desired output canvas dimensions [width, height]. Default [1000, 1000]. Max 5000x5000 pixels.	`—`	—
`content_moderation`	boolean	Enable content moderation	`false`	—
`image`	string(uri)	Image file	`—`	—
`image_url`	string	Image URL	`—`	—
`negative_prompt`	string	Negative prompt for image generation	`—`	—
`original_image_location`	array	Position of original image in canvas [x, y]	`—`	—
`original_image_size`	array	Size of original image in canvas [width, height]	`—`	—
`preserve_alpha`	boolean	Preserve alpha channel in output. When true, maintains original transparency. When false, output is fully opaque.	`true`	—
`prompt`	string	Text prompt for image generation	`—`	—
`seed`	integer	Random seed. Set for reproducible generation	`—`	—
`sync`	boolean	Synchronous response mode	`true`	—

aspect_ratiostring

Aspect ratio for expansion. Either aspect_ratio or canvas_size with original_image_size/location must be provided. Can be a predefined string like '1:1', '16:9' etc. or a custom float between 0.5 and 3.0

Default: "1:1"

1:12:33:23:44:34:55:49:1616:9

canvas_sizearray

Desired output canvas dimensions [width, height]. Default [1000, 1000]. Max 5000x5000 pixels.

content_moderationboolean

Enable content moderation

Default: false

imagestring

Image file

image_urlstring

Image URL

negative_promptstring

Negative prompt for image generation

original_image_locationarray

Position of original image in canvas [x, y]

original_image_sizearray

Size of original image in canvas [width, height]

preserve_alphaboolean

Preserve alpha channel in output. When true, maintains original transparency. When false, output is fully opaque.

Default: true

promptstring

Text prompt for image generation

seedinteger

Random seed. Set for reproducible generation

syncboolean

Synchronous response mode

Default: true

Version: 0d8d951a482dUpdated: 7/25/2026350.7K runs