GET

/v1/schemas/bfl/image/v1/flux-tools/outpainting-v1

Accept:

kind inputactivity imagehash 8f94b5d8…extracted Oct 22, 2026version currentsource upstream spec ↗

field	recorded source
schema	Source linked above. upstream-spec Source trace `ea458c661ef8b0e03d72ee6b29267451f5a04ae65f35e55ea8a87e53645ef670` { "fetchedAt": 1792645846, "extractorVersion": "1", "verifiedAt": null }

Request body

12 properties · 3 required · 1 $defs

property	type	description	constraints
input_image*	string	Base64-encoded input (reference) image or HTTP(S) image URL.
width*	integer	Target output width.	n ≥ 64
height*	integer	Target output height.	n ≥ 64
user	string	Opaque identifier for the end user supplied by the calling platform.	length ≥ 1 · length ≤ 256 · nullable
auto_crop	boolean	If True, crop the input image to the canvas bounds when it extends beyond the edges. If False, an error is raised instead.	default: false
safety_tolerance	integer	Harm moderation tolerance from 0 to 7, with 0 the strictest. The route and organization determine the approved maximum and per-category ceilings. Unapproved levels return 422 with approved levels.	0 ≤ n ≤ 7 · default: 2
output_format	enum		jpeg \| png \| webp · nullable
prompt	string	Experimental: optional text guidance for the outpainted region. The model may not strictly follow this prompt; the visual content of the input image is the primary signal. Leave unset for default behavior.	nullable
reference_offset_x	integer	Left offset (px) of the reference image's top-left corner on the output canvas. Negative values are allowed. None = center horizontally.	nullable
reference_offset_y	integer	Top offset (px) of the reference image's top-left corner on the output canvas. Negative values are allowed. None = center vertically.	nullable
mode	enum	Quality/speed trade-off. 'high' (default): highest-fidelity results, recommended whenever fine detail, prompt adherence, or consistency with complex content in the source image matters; slower. 'fast': significantly faster and well-suited for naturally extendi…	high \| fast · default: "high"
disable_pup	boolean	Skip the image-aware prompt upsampler for lower latency. The prompt (or a default extension prompt) is sent directly to the generation model; quality in the extended region may be lower for scenes that benefit from semantic guidance. Applies to both modes.	default: false

Types

1 $defs

OutputFormatenum

jpeg | png | webp

Validate a payload

before you spend tokens

shell

$ curl -X POST https://modelschemas.com/v1/validate \
    -d '{"provider":"bfl","endpointId":"v1/flux-tools/outpainting-v1","payload":{…}}'
→ {"valid": false, "errors": [{"path": "#", "keyword": "required", …}]}