ComfyUI Node

Ideogram 2a

Ideogram 2a is the lean cloud option for text-on-image

By Runware·Created 2 years ago·Updated about a month ago· 140
Ideogram 2a
  • advancedFeatures.watermark.image
  • providerSettings.ideogram.styleReferenceImages
  • image
positivePrompt
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
providerSettings.ideogram.magicPromptAUTO
safetyfalse
safety.checkContentfalse
providerSettings.ideogram.styleTypeAUTO
ttlfalse
ttl_value60
outputFormatJPG
outputQuality95

Ideogram 2a is the budget text-to-image tier of the lab that made accurate text-in-image a solved problem. Where the Remix nodes need a source image and the 3.0 node needs a plan, 2a is just: prompt, size, go. It's the node you reach for when a poster, a logo draft, or a sign with legible lettering has to exist right now and your local SDXL checkpoint is still spelling "COFFEE" as "COFEE."

It runs through the Runware/ComfyUI-Runware pack, so the whole thing is one API call away - no local model files, no GPU requirements, and the output is a native IMAGE tensor you can feed straight into a Save Image node or the next step of a workflow.

How it works

The node sends your prompt plus dimensions to Ideogram's 2a model via Runware's Python SDK over REST. Ideogram's whole research lineage (ex-Google Brain, the people behind DDPM and Imagen, per the KB) is why the lettering comes out right: the model was built around text rendering, not as an afterthought. You get the model's parameters as widgets - there's no sampler or step count to babysit.

The inputs that matter

Honestly, there are only a handful you'll touch:

  • positivePrompt - required. Put the text you want in quotes in the prompt, e.g. neon sign reading "OPEN 24 HOURS", night street.
  • width / height - required, 1024 default.
  • providerSettings.ideogram.styleType - AUTO, GENERAL, REALISTIC, DESIGN, or ANIME. DESIGN biases toward clean graphic-design output; REALISTIC for photographs.
  • providerSettings.ideogram.magicPrompt - AUTO enhances your prompt automatically. It's often good; it's also the reason your carefully-worded prompt comes back reworded. Set OFF when exact wording matters.
  • providerSettings.ideogram.styleReferenceImages - optional style image guidance, which 2a supports even though it's the lightweight tier.

numberResults (up to 20), seed, watermark, output format/quality, and the standard safety/TTL fields fill out the rest.

Installing

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

Restart ComfyUI, or install via ComfyUI Manager by searching Runware. API key via Settings → Runware API key, the RUNWARE_API_KEY env var, or runware auth login.

Common issues

The classic complaint is "it rewrote my text" - that's magicPrompt on AUTO, so turn it off if your exact lettering matters. And while 2a is the cheap tier, "cheap" is relative to the other Ideogram tiers; the title bar on the node shows each run's cost, so watch it if you're doing 20-result batches. Ideogram's built-in content filtering applies here too - a refused prompt is the model's call, not a node error. If you need character consistency or 63 style presets, step up to the 3.0 node instead; this one is for fast, clean text.

CategoryRunware/Image/ideogram

Inputs (20)

NameTypeDefaultDescription
positivePromptSTRINGText prompt describing elements to include in the generated output.
advancedFeatures.watermark.imageoptIMAGE
providerSettings.ideogram.styleReferenceImagesoptIMAGE
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.
providerSettings.ideogram.magicPromptoptCOMBOAUTOAutomatic prompt enhancement.
safetyoptBOOLEANfalseEnable to set safety. Off uses the model's default.
safety.checkContentoptBOOLEANfalseEnable or disable content safety checking.
providerSettings.ideogram.styleTypeoptCOMBOAUTOVisual style and rendering approach.
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