Skip to main content
Recraft V3, released in October 2024, is a previous-generation model known for photorealism and text rendering. In the API it has the widest operation support: inpainting, outpainting, and the background operations run on V3 only, and it is the only line that places text at exact positions.

$0.04

per image, raster

1K

WebP, PNG, or SVG

1,000

characters of prompt

recraftv3

model value

Try it

Get an API token on your profile page (it needs a positive API units balance) and set it once per terminal session:
Then run:

Response

  • data has one entry per image (n). Download each image from its url.
  • Image URLs are temporary: files are kept for about 24 hours, so download them. See Limits and sizes.
  • credits is the number of API units charged for the request.
  • Pass "response_format": "b64_json" to get the image bytes inline, or "multipart" for the lowest latency. See Image inputs and results.

Parameters

Variants

Capabilities

Also supported: curated and custom styles, negative_prompt, text layout, and the full set of controls.

Examples

Each bbox is four [x, y] points relative to the image, from (0, 0) top-left to (1, 1) bottom-right.
Regenerate the white area of mask.png in image.png:
See Image inpainting for mask requirements.

Curated styles

The full list of V3 and V3 Vector styles.

Endpoints

Full request reference for generation and every editing operation.

All models

The whole lineup with prices and best-fit use cases.

About Recraft V3

The model’s background and strengths in detail.