ComfyUI Node

FLUX.1 Expand [pro]

Growing your canvas, credibly

By Runware·Created 2 years ago·Updated about a month ago· 140
FLUX.1 Expand [pro]
  • seedImage
  • outpaint
  • advancedFeatures.watermark.image
  • image
positivePrompt
steps50
CFGScale60.00
numberResults1
advancedFeaturesfalse
advancedFeatures.watermark.bgColor
advancedFeatures.watermark.displayPosition(default)
advancedFeatures.watermark.fontColor
advancedFeatures.watermark.opacityfalse
advancedFeatures.watermark.opacity_value0.10
advancedFeatures.watermark.text
providerSettings.bfl.promptUpsamplingfalse
safetyfalse
safety.checkContentfalse
providerSettings.bfl.safetyTolerance2
ttlfalse
ttl_value60
outputFormatJPG
outputQuality95

Outpainting is the inverse of inpainting: instead of filling a hole, you extend the image past its edges and let the model invent the scenery that continues. Done badly, you get smeared gradients where the old canvas ends. FLUX.1 Expand [pro] is BFL's dedicated outpainting model - API-only - and it's genuinely good at extending a photo or illustration so the seam is hard to find.

This node is the Runware wrapper for it. The outpaint input is a typed socket you feed from a Runware/Params builder, which is where you say which direction(s) to expand and by how much. That builder-driven design is the one thing that trips people up, so let's walk it.

How it works

A taskType: imageInference REST call (model bfl:1@3) with your source image uploaded as a data URI. The outpaint builder's output gets merged into the request as the expansion geometry, and the model generates the new pixels beyond the original canvas, guided by your prompt. Result comes back as a native IMAGE - usually larger than the input because the canvas itself grew.

Inputs that matter

  • seedImage - the image to expand. Required.
  • outpaint - the builder socket. This is not a number you type on the node; it's the output of the outpaint param node, where you configure direction (left/right/up/down) and extent. If you haven't used a Runware builder before, find it under Runware/Params.
  • positivePrompt - what the extended region should contain ("continue the same beach, waves on the right").
  • steps - 15–50, default 50. BFL's expand model is a full-step model; the default 50 is on the high end, so you can trim it if cost matters.
  • CFGScale - 1.5–100, default 60. That default looks wild compared to FLUX norms, but this model is tuned for high guidance. Leave it unless you know why you're changing it.

Install

cd ComfyUI/custom_nodes
git clone https://github.com/Runware/ComfyUI-Runware
pip install -r ComfyUI-Runware/requirements.txt

Or ComfyUI Manager → search Runware → Install → restart. API key in ComfyUI Settings → Runware API key or RUNWARE_API_KEY.

Gotchas

  • Forgetting the outpaint builder is the classic miss - the node won't run without it, and it's easy to mistake the socket for an optional extra. It's required.
  • Outpainting large amounts in one call is where quality degrades. If you're doubling a canvas, consider two passes with a mid-step upscale rather than one giant jump.
  • The image output is bigger than your input; if you're chaining into other nodes, check your downstream resolution assumptions.

Expand is the specialist's tool in this pack - most people reach for the general FLUX nodes, but when you need a believable, edge-consistent canvas extension, this is the one. Just remember the builder.

CategoryRunware/Image/bfl

Inputs (22)

NameTypeDefaultDescription
seedImageIMAGE
outpaintRUNWARE_OUTPAINT
positivePromptSTRINGText prompt describing elements to include in the generated output.
advancedFeatures.watermark.imageoptIMAGE
stepsoptINT5015–50Total number of denoising steps. Higher values generally produce more detailed results but take longer.
CFGScaleoptFLOAT60.001.5–100Guidance scale representing how closely the output will resemble the prompt. Higher values produce results more aligned with the prompt.
numberResultsoptINT11–20Number of results to generate. Each result uses a different seed, producing variations of the same parameters.
advancedFeaturesoptBOOLEANfalseEnable to set advancedFeatures. Off uses the model's default.
advancedFeatures.watermark.bgColoroptSTRINGBackground color in hex format.
advancedFeatures.watermark.displayPositionoptCOMBO(default)Watermark position.
advancedFeatures.watermark.fontColoroptSTRINGText color in hex format.
advancedFeatures.watermark.opacityoptBOOLEANfalseEnable to set advancedFeatures.watermark.opacity. Off uses the model's default.
advancedFeatures.watermark.opacity_valueoptFLOAT0.100.1–1Watermark opacity.
advancedFeatures.watermark.textoptSTRINGWatermark text.
providerSettings.bfl.promptUpsamplingoptBOOLEANfalseEnable automatic enhancement and expansion of the input prompt.
safetyoptBOOLEANfalseEnable to set safety. Off uses the model's default.
safety.checkContentoptBOOLEANfalseEnable or disable content safety checking.
providerSettings.bfl.safetyToleranceoptINT20–5Tolerance level for content moderation. Lower values are stricter; higher values are more permissive.
ttloptBOOLEANfalseEnable to set ttl. Off uses the model's default.
ttl_valueoptINT60Time-to-live (TTL) in seconds for generated content. Only applies when `outputType` is `URL`.
outputFormatoptCOMBOJPGFile format for the generated image.
outputQualityoptINT9520–99Compression quality of the output. Higher values preserve quality but increase file size.

Outputs (1)

NameTypeDescription
imageIMAGE