ComfyUI Node

Bria FIBO Lite

Bria FIBO Lite — the version of FIBO that took away your dials

By Runware·Created 2 years ago·Updated about a month ago· 140
Bria FIBO Lite
  • image
  • advancedFeatures.watermark.image
  • image
positivePrompt
negativePrompt
size1K (1:1)
seed1
CFGScale5
numberResults1
advancedFeaturesfalse
advancedFeatures.watermark.bgColor
advancedFeatures.watermark.displayPosition(default)
advancedFeatures.watermark.fontColor
advancedFeatures.watermark.opacityfalse
advancedFeatures.watermark.opacity_value0.10
advancedFeatures.watermark.text
providerSettings.bria.ipSignalfalse
safetyfalse
safety.checkContentfalse
ttlfalse
ttl_value60
outputFormatJPG
outputQuality95

Compare this node to the full Bria FIBO node and the first thing you'll notice is everything that's missing. No steps. No seed. No negativePrompt. That's not a bug - FIBO Lite is the fast sibling of FIBO (it shares the same model family, just a later, leaner version), and the pack has deliberately removed the knobs you'd use to slow yourself down. You get a prompt, dimensions, an optional input image, the three-value CFG choice, and results. That's the whole node.

There's a real argument for this. FIBO's full node hands you steps 20–50, guidance 3/4/5, seeds, and negatives - and on a model that's already trained for speed, half of those are ways to accidentally produce worse output. Lite is the "just give me the image" variant: fire and forget, which makes it the one you'll reach for in draft mode or when you're batch-testing prompts and don't want 50-step waits (or 50-step bills) on every sketch.

What's actually here

  • positivePrompt - required, multiline. Same long-caption-friendly behavior as the full model.
  • width / height - required, default 1024×1024.
  • image - optional IMAGE input for starting from an existing image.
  • CFGScale - the only guidance knob, and it's an enum of 3/4/5. FIBO runs best at low CFG; the node just won't let you leave that range.
  • numberResults - 1 to 20. Each result is a separate paid inference with a fresh seed. The one cost lever, and the one to be careful with.

Output is a single IMAGE socket into Preview or Save. Everything else (outputFormat JPG/PNG/WEBP, outputQuality, safety, the ipSignal flag) is optional and safe to leave at defaults.

Install and running

Same one-time pack setup, no model downloads because inference happens on Runware's GPUs:

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

Restart, then drop your Runware API key into Settings → Runware API key (or set RUNWARE_API_KEY). The node is a thin client: your prompt and image tensor go up as a request, an IMAGE tensor comes back, and the run's cost shows on the title bar.

The honest trade

What you're giving up by picking Lite is control, not just clutter. No seed means no reproducible batches - each run is a different result, so if you land on something you love you can't lock it down. No negative prompt means the only way to push the output away from something is to phrase your positive prompt better. And same as the full FIBO nodes, the weights are non-commercial licensed, so check the terms before this becomes part of a product. For personal drafts and quick ideation, none of that registers - Lite is the one you'll actually keep on the canvas.

CategoryRunware/Image/bria

Inputs (22)

NameTypeDefaultDescription
positivePromptSTRINGText prompt describing elements to include in the generated output.
imageoptIMAGE
advancedFeatures.watermark.imageoptIMAGE
negativePromptoptSTRINGPrompt to guide what to exclude from generation. Ignored when guidance is disabled (CFGScale ≤ 1).
sizeoptCOMBO1K (1:1)Output size. Only one of resolution or dimensions is sent.
seedoptINT11–2147483647Random seed for reproducible generation. When not provided, a random seed is generated in the unsigned 32-bit range.
CFGScaleoptCOMBO5Guidance 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.bria.ipSignaloptBOOLEANfalseFlag potential IP-related content in prompt or output.
safetyoptBOOLEANfalseEnable to set safety. Off uses the model's default.
safety.checkContentoptBOOLEANfalseEnable or disable content safety checking.
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