1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
const axios = require('axios');
const FormData = require('form-data');
const api_key = "YOUR API-KEY";
const url = "https://api.segmind.com/v1/ideogram-character";
const reqBody = {
"prompt": "a close up photo of a woman in a fashion magazine photoshoot",
"resolution": "512x1536",
"style_type": "Realistic",
"aspect_ratio": "16:9",
"rendering_speed": "Quality",
"magic_prompt_option": "Off",
"character_reference_image": "https://segmind-resources.s3.amazonaws.com/output/9ad9b775-7b35-4485-affe-6215a933d825-686185927069985948.png"
};
(async function() {
try {
const formData = new FormData();
// Append regular fields
for (const key in reqBody) {
if (reqBody.hasOwnProperty(key)) {
formData.append(key, reqBody[key]);
}
}
// Convert and append images as Base64 if necessary
const response = await axios.post(url, formData, {
headers: {
'x-api-key': api_key,
...formData.getHeaders()
}
});
console.log(response.data);
} catch (error) {
console.error('Error:', error.response ? error.response.data : error.message);
}
})();Attributes
Use a mask to preserve parts of the image while inpainting. Choose a detailed mask for precise edits.
Set a seed for reproducible results. Use a fixed seed for consistency across generations.
Provide an image for inpainting using the given mask. Start with a clear, high-resolution image for best results.
Describe the image you want to create. Use creative prompts for unique images.
Choose resolution for the image output. For detailed images, pick a higher resolution.
Allowed values:
Select the style for the image. Use 'Fiction' for creative artwork or 'Realistic' for lifelike images.
Allowed values:
Define the aspect ratio for your image. Choose a ratio like '16:9' for wide-screen images.
Allowed values:
Choose rendering speed for processing time. Use 'Turbo' for fast results or 'Quality' for best results.
Allowed values:
Toggle magic prompt modes to automate text prompts. Use 'On' for automatic prompt enhancements.
Allowed values:
Provide an image as a character reference. Use a clear headshot for accurate modeling.
To keep track of your credit usage, you can inspect the response headers of each API call. The x-remaining-credits property will indicate the number of remaining credits in your account. Ensure you monitor this value to avoid any disruptions in your API usage.
Ideogram Character - AI Character Consistency Model
Edited by Segmind Team on September 18, 2025.
What is Ideogram Character?
Ideogram Character is an AI image model designed to create multiple near-perfect character representations using only one image as the reference prompt. The model maintains the original subject's accurate visual reproduction with high-resolution character variations. It can be used to design creatives for actual individuals or fictional characters to develop scalable headshots or portrayals, which are uniform in facial structure, expressions, and even stylistic accuracy.
Additionally, Ideogram Character is easily compatible with direct platform use as well as API access, making it a popular and reliable option for creative workflows and tasks that need large-scale character generation outputs in a short time.
Key Features of Ideogram Character
- Single-reference consistency - It can maintain character accuracy using only one input image
- Dual rendering modes - You can select between "Turbo" for speed or "Quality" for maximum fidelity
- Flexible aspect ratios - It supports 25+ aspect ratios from 1:3 to 16:9
- Style control - It provides either "Realistic" or "Fiction" rendering styles
- Magic prompt enhancement - It can automate prompt optimization for better results
- Inpainting capabilities - It has mask-based controls to edit specific image regions
- Reproducible outputs - It has the option to set specific seed parameters for consistent outputs
Best Use Cases
- Content Creation - You can create character variations for storytelling, comics, and digital narratives without relying on multiple models or artists.
- Game Development - Designers can maintain the character recognition by creating consistent NPC appearances across different scenes and contexts.
- Marketing & Advertising - Creative teams can produce diverse headshots and character representations for their campaigns to ensure consistent brand personas.
- E-Learning & Training - Educators can develop multiple modules based on character-driven educational content with high-end visual consistency.
Prompt Tips and Output Quality
- Write descriptive prompts focusing on scene, lighting, and context instead of facial features. For example, "professional headshot in office setting, soft lighting" works better than describing facial characteristics.
- Begin with the magic_prompt_option set to "On" for automatic prompt enhancement.
- The style_type parameter significantly impacts output: choose "Realistic" for photographic quality or "Fiction" for artistic interpretations.
- Higher resolutions (1024x1024+) provide better detail retention, especially for facial features.
- Set rendering_speed to "Quality" for final outputs and "Turbo" for rapid prototyping.
FAQs
Is Ideogram Character open-source?
No, Ideogram Character is a proprietary model available through Ideogram's platform and API services.
How does it differ from other character generation models?
Ideogram Character specializes in maintaining perfect character consistency across generations using minimal reference input.
What image formats work best for character references?
Clear headshots in common formats (JPEG, PNG, WebP) with good lighting and facial visibility produce optimal results.
What resolution should I use for production work?
For professional applications, use the resolution '1024x1024' or higher.
The model supports resolutions up to 1536x640; you can use it as per your desired aspect ratio.
Can I control facial expressions and poses?
Yes, describe desired expressions and poses in your text prompt; the model will maintain the original character features based on the reference prompt.
How do I achieve reproducible results?
Set a specific seed value in the parameters since using the same seed with identical prompts and settings will generate consistent outputs.
Popular Models
