ComfyUI Node

Exactly Warm Light

Golden-hour vibes, from a node with no prompt

By Runware·Created 2 years ago·Updated about a month ago· 140
Exactly Warm Light
  • referenceImages
  • advancedFeatures.watermark.image
  • image
size1K (1:1)
numberResults1
advancedFeaturesfalse
advancedFeatures.watermark.bgColor
advancedFeatures.watermark.displayPosition(default)
advancedFeatures.watermark.fontColor
advancedFeatures.watermark.opacityfalse
advancedFeatures.watermark.opacity_value0.10
advancedFeatures.watermark.text
safetyfalse
safety.checkContentfalse
ttlfalse
ttl_value60
outputFormatJPG
outputQuality95

Warm Light is the Exactly photo node that speaks golden hour. The name is the whole spec: warm, low, flattering illumination as the default tone of every render. It's the variant you grab when the brief calls for coziness, portraiture, or anything that should feel lit by the end of the day - and like the rest of its family, it takes no text prompt at all. Your references define the subject and identity; the model supplies the warmth.

That's the family trick, and it's worth repeating because it's the part people don't expect: there is no prompt field on these nodes at all. You curate references that establish who and what, then pick the variant that sets the mood. Warm Light is the mood option for everyone who's ever typed "golden hour, warm tones" a thousand times into a prompt box and wished it were automatic.

What you set

  • referenceImages (required) - from the Runware Reference Images builder: each ref takes a type, a tag (like @subject) and the image. Warm-toned references will reinforce the look; cool ones will fight it a little.
  • width / height - output canvas, default 1024×1024.
  • numberResults - up to 20 seeded variations.
  • watermark - optional builder socket.
  • outputFormat - JPG / PNG / WEBP, outputQuality 20–99.

Output is image (IMAGE).

How it works

The standard pack mechanism: a taskType: imageInference request carrying the Exactly Warm Light model AIR goes over REST via the Runware SDK, your reference images ride along in the payload, and the rendered image comes back as a native IMAGE. All the inference happens on Runware's hardware - no checkpoint download, no local VRAM, and the title bar tells you the per-run cost. The warm bias is baked into the hosted model, so consistency across a batch is a property of the API, not of your skill at prompt taming.

Installing

ComfyUI Manager → search Runware → install → restart. Manual:

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

No model downloads; deps are runware-sdk, pillow, soundfile. API key via Settings → Runware API key, RUNWARE_API_KEY, or runware auth login.

Troubleshooting

If the warmth isn't showing, check your references aren't dragging the palette cool - the model's bias is strong but not absolute. Faces coming out too orange? That's the warm tone over-applied; cooler, more neutral references tame it. And when a project needs neutral light instead of golden, Exactly Journey is the no-bias baseline to fall back on.

CategoryRunware/Image/exactly

Inputs (17)

NameTypeDefaultDescription
referenceImagesRUNWARE_REFERENCEIMAGES
advancedFeatures.watermark.imageoptIMAGE
sizeoptCOMBO1K (1:1)Output size. Only one of resolution or dimensions is sent.
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.
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