Skip to main content
ModelRunner
ideogram avatar

ideogram / character

Generate new images of the same character from one reference photo and a text prompt, keeping facial features and distinctive traits consistent across scenes.

0.15
OpenAPI

Model Input

Input

The text prompt describing the image to generate. The character's identity comes from the reference image; describe the scene, pose, wardrobe, and style here. Put any words you want rendered inside the image in quotes.

  • https://media.modelrunner.ai/FYKMceHwpZ2cR52xlL2gx.png

A set of images to use as character references. Currently only the first image is used; the rest are ignored (maximum total size 10MB). JPEG, PNG, or WebP.

The speed/quality/cost tradeoff. TURBO is fastest and cheapest, BALANCED is the default, QUALITY is best but slowest and most expensive. Affects both speed and cost.

The style type to generate with. AUTO lets the model choose, REALISTIC biases toward photographic results, FICTION toward stylized/illustrative results. Cannot be used with style_codes.

Additional Settings

Customize your input with more control.

If true, automatically enriches the prompt with extra descriptive detail before generation (Magic Prompt). Set false to use your prompt exactly as written.

The size of the generated image. Use a preset string (e.g. 'landscape_16_9') or a custom {width, height} object.

The same seed and the same prompt given to the same version of the model will output the same image every time.

A description of what you do NOT want in the generated image.

Optional masks for the character reference images, restricting which regions are used as references.

Optional style reference images (distinct from the character references). Cannot be used together with style_codes.

Optional 8-character hex style codes. Cannot be used together with style or style reference images.

You need to be logged in to run this model and view results.
Log in

Model Output

Output

Generated image output
Generated in 18.493 seconds
Logs (1 lines)

Model Example Requests

Examples

Example output 1Example output 2

Model Details

Model Details

Ideogram Character takes a single reference image of a person or character plus a text prompt and generates new images that keep that character's identity consistent — the same face, proportions, and distinctive traits — placed into whatever scene, pose, or style the prompt describes. Provide one clear reference portrait and describe the new shot, and it renders the same character on a rooftop, in a comic panel, in a product ad, or across a story sequence without redrawing them from scratch each time. Its standout strength is character consistency: unlike a plain text-to-image model, the output stays recognizably the same individual from image to image.

## Best for - Character-consistent storytelling: comics, storyboards, and illustrated narratives that reuse the same protagonist across panels - Branding and mascots that must look identical across a campaign or asset set - Placing one person or character into new scenes, outfits, poses, or lighting from a single reference photo - Producing on-model or in-scene images that all read as the same individual across a series of runs

## Choose another model when - You want to restyle or edit the reference image itself while keeping its composition — use an image-to-image editing model - You need generation from a text prompt alone with no character to preserve — use a text-to-image model - You need to preserve a scene or object rather than a person's identity — this model is tuned for character/subject consistency - You need the character animated in motion — use an image-to-video model

## Tips - Provide one clear, high-resolution reference image; only the first `reference_image_urls` entry is used, so extra references are ignored - Describe the new scene, pose, wardrobe, and mood in the prompt — the reference supplies identity, the prompt supplies everything else - Use `rendering_speed` to trade cost and speed against quality: TURBO for quick drafts, QUALITY for final hero images - Put any words you want rendered inside the image in quotes in the prompt

## Advanced Configuration

### rendering_speed Controls the speed/quality/cost tradeoff. `TURBO` is fastest and cheapest, `BALANCED` (the default) is the middle ground, and `QUALITY` is best but slowest. This affects both latency and per-image cost.

### style Sets the overall look: `AUTO` lets the model choose, `REALISTIC` biases toward photographic results, and `FICTION` biases toward stylized/illustrative results. Cannot be combined with `style_codes`.

To run via the ModelRunner JavaScript client: ```js import { modelrunner } from "@modelrunner/client";

const result = await modelrunner.subscribe("ideogram/character", { input: { prompt: "the same woman standing on a city rooftop at golden hour, wind in her hair, cinematic", reference_image_urls: ["https://media.modelrunner.ai/example-portrait.png"], rendering_speed: "BALANCED", style: "REALISTIC", }, }); ```