ComfyUI Node

FLUX.1.1 [pro]

BFL's closed flagship, on your canvas

By Runware·Created 2 years ago·Updated about a month ago· 140
FLUX.1.1 [pro]
  • advancedFeatures.watermark.image
  • image
positivePrompt
width1024
height1024
seed0
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

FLUX.1.1 [pro] is the model you literally cannot run yourself. Black Forest Labs' flagship never left the API - the weights were never released, which was the entire point. It's the "we keep the good stuff behind the paywall" tier of the FLUX family, roughly triple the speed of FLUX.1 pro and significantly better at following complex, multi-part prompts. The only way to touch it at all is through someone's cloud. This node is that someone's cloud, glued into ComfyUI.

That's the pitch for every BFL node in this pack: they wrap the closed pro-tier API models as first-class nodes, so you get the proprietary quality without leaving your graph. Worth it for prompt adherence alone - FLUX.1.1's guidance-following is a visible step above the open dev/schnell tiers.

How it works

A thin taskType: imageInference request over Runware's REST API, model AIR bfl:2@1. Widget values get packed into the request, Runware's fleet runs the closed model, and the node downloads the result to an IMAGE tensor. Nothing about this model's internals is documented - you're trusting BFL's servers, full stop.

Inputs that matter

  • positivePrompt - required. FLUX.1.1 is a natural-language model; write sentences, not tag soup.
  • width / height - 256–1440 in 32px steps. BFL's pro tier has a hard-ish resolution ceiling, so don't fight it - upscale afterward instead.
  • providerSettings.bfl.promptUpsampling - the auto-prompt-expander. Off by default; it's worth flipping on for short prompts.
  • providerSettings.bfl.safetyTolerance - 0–5, default 2. This is BFL's content moderation dial: lower is stricter, higher is more permissive. If your (legitimate) subject keeps getting flagged, this is the knob.
  • seed - set it for reproducibility; the default randomizes per run.
  • numberResults - batch variations cheaply, each with its own seed.

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 the RUNWARE_API_KEY env var. Prepaid top-up; per-run cost shows on the node title bar.

Gotchas

  • It's the pro tier, so it's the priciest per image of the FLUX family. Great results, but budget for it - this isn't the node for mass iteration.
  • Moderation is real and lives on the provider side; safetyTolerance is your only lever, and it tops out at 5.
  • No steps widget here - BFL fixes the step count on their end. If you want control, the dev-tier FLUX nodes in this pack give it to you; pro trades that away for turnkey quality.

If you need BFL's closed model in a ComfyUI workflow and can't be bothered with yet another API client, this node is the whole integration. The other FLUX.1.1 flavor in the pack - [pro] Ultra - is the faster, even more photoreal sibling.

CategoryRunware/Image/bfl

Inputs (21)

NameTypeDefaultDescription
positivePromptSTRINGText prompt describing elements to include in the generated output.
widthINT1024256–1440Width of the generated media in pixels.
heightINT1024256–1440Height of the generated media in pixels.
advancedFeatures.watermark.imageoptIMAGE
seedoptINT00–9223372036854776000Random seed for reproducible generation. When not provided, a random seed is generated in the unsigned 32-bit range.
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