IdeogramTextToImageApi - ComfyUI Built-in Node Documentation

Generate images from a text prompt with Ideogram 4.5. The prompt also accepts an Ideogram structured JSON caption, for example the final_prompt returned by an earlier run, which gives exact control over text strings, colors and layout. The node returns the generated image(s) together with the caption the image was actually generated from.

Inputs

Parameter	Description	Data Type	Required	Range
`model`	Model to use. (default: `"ideogram-4.5"`)	DYNAMIC_COMBO	Yes	`"ideogram-4.5"`
`prompt`	Text prompt, or an Ideogram structured JSON caption such as a previous `final_prompt`. (default: empty string)	STRING	Yes	1 to 10000 characters
`size`	Output size. `"auto"` lets the model pick a canvas that suits the prompt. A `(2K)` or `(1K)` preset fixes both the pixel size and the aspect ratio. (default: `"auto"`)	COMBO	Yes	`"auto"` `"(2K) 2048x2048 (1:1)"` `"(2K) 1440x2880 (1:2)"` `"(2K) 2880x1440 (2:1)"` `"(2K) 1664x2496 (2:3)"` `"(2K) 2496x1664 (3:2)"` `"(2K) 1792x2240 (4:5)"` `"(2K) 2240x1792 (5:4)"` `"(2K) 1440x2560 (9:16)"` `"(2K) 2560x1440 (16:9)"` `"(2K) 1600x2560 (5:8)"` `"(2K) 2560x1600 (8:5)"` `"(2K) 1728x2304 (3:4)"` `"(2K) 2304x1728 (4:3)"` `"(2K) 1296x3168 (9:22)"` `"(2K) 3168x1296 (22:9)"` `"(2K) 1152x2944 (9:23)"` `"(2K) 2944x1152 (23:9)"` `"(2K) 1248x3328 (3:8)"` `"(2K) 3328x1248 (8:3)"` `"(2K) 1280x3072 (5:12)"` `"(2K) 3072x1280 (12:5)"` `"(2K) 1024x3072 (1:3)"` `"(2K) 3072x1024 (3:1)"` `"(1K) 1024x1024 (1:1)"` `"(1K) 896x1120 (4:5)"` `"(1K) 1120x896 (5:4)"` `"(1K) 864x1152 (3:4)"` `"(1K) 1152x864 (4:3)"` `"(1K) 832x1248 (2:3)"` `"(1K) 1248x832 (3:2)"` `"(1K) 800x1280 (5:8)"` `"(1K) 1280x800 (8:5)"` `"(1K) 720x1280 (9:16)"` `"(1K) 1280x720 (16:9)"` `"(1K) 720x1440 (1:2)"` `"(1K) 1440x720 (2:1)"`
`quality`	Quality tier. Higher tiers cost more and take longer. (default: `"medium"`)	COMBO	Yes	`"low"` `"medium"` `"high"`
`magic_prompt`	Rewrites the prompt into a detailed structured caption before generating; `"off"` keeps your wording as literal as possible. The caption is returned as `final_prompt`. (default: `"auto"`) This is an advanced setting.	COMBO	Yes	`"auto"` `"on"` `"off"`
`seed`	Seed for generation. Text-to-image is not reproducible from the seed alone because the prompt is rewritten on every run; to reproduce an image, reuse its `final_prompt` with `magic_prompt` set to `"off"` and the same seed. (default: 42)	INT	Yes	0 to 2147483647

Parameter Constraints

Prompt required: the prompt must contain at least one non-whitespace character and at most 10000 characters. Set magic_prompt to "off" when supplying your own JSON caption or exact wording.
Size: "auto" lets the model choose. (2K) presets are about 3 to 4 megapixels and (1K) presets about 1 megapixel, so the tier label describes the pixel budget rather than a fixed long side; the preset’s aspect ratio is respected. Only the pixel size part of the preset is sent to the API.
Reproducibility: with magic_prompt set to "auto" or "on" the prompt is rewritten on every run, so the same seed can still produce a different image. To reproduce an image, feed its final_prompt back with magic_prompt set to "off" and the same seed.
Content safety: if Ideogram’s content safety filter blocks the generation, the node raises an error instead of returning an image.

Outputs

Output Name	Description	Data Type
`IMAGE`	The generated image(s) as a batch.	IMAGE
`final_prompt`	The structured caption the image was generated from. Feed it back with `magic_prompt` set to `"off"` and the same seed to reproduce the image.	STRING

This documentation was AI-generated. If you find any errors or have suggestions for improvement, please feel free to contribute! Edit on GitHub

Source fingerprint (SHA-256): a21f1faed9ca7a7bc63dae74003cc7599f11013075f718af9e5860d2cd666828

​Inputs

​Parameter Constraints

​Outputs

Inputs

Parameter Constraints

Outputs