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
const axios = require('axios');
const fs = require('fs');
const path = require('path');
async function toB64(imgPath) {
const data = fs.readFileSync(path.resolve(imgPath));
return Buffer.from(data).toString('base64');
}
const api_key = "YOUR API-KEY";
const url = "https://api.segmind.com/v1/qwen-image-edit-plus-add-people";
const data = {
"prompt": "Generate a cozy family sitting on the sofa and a dog on the floor",
"image_1": "toB64('https://segmind-resources.s3.amazonaws.com/input/4ef80f21-c076-445f-b0cc-af37e2619733-cozy-living-room-sofa-against-an-empty-white-brick-wall-modern-home-decor-indoor-setting-minimalist-style-free-photo.jpg')",
"image_2": "toB64('')",
"image_3": "toB64('')",
"lora": "add_people",
"aspect_ratio": "4:3",
"seed": 87568756,
"image_format": "webp",
"quality": 95,
"base64": false
};
(async function() {
try {
const response = await axios.post(url, data, { headers: { 'x-api-key': api_key } });
console.log(response.data);
} catch (error) {
console.error('Error:', error.response.data);
}
})();Attributes
Text prompt describing the desired image edit or generation
Primary input image (URL or base64)
Secondary input image (URL or base64)
Tertiary input image (URL or base64)
Pre-configured LoRA model to apply
Allowed values:
Additional LoRA model URL. Public direct url or huggingface url pointing to the lora file
Additional LoRA model URL. Public direct url or huggingface url pointing to the lora file
Output image aspect ratio
Allowed values:
Random seed for reproducibility. Use -1 for random
min : -1,
max : 2147483647
Output image format
Allowed values:
Output image quality (1-100)
min : 1,
max : 100
Return image as base64 encoded string
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.
Add people to the scene: Image-to-Image Model
Edited by Segmind Team on November 27, 2025.
What is Qwen Image Edit Plus Add People?
Qwen Image Edit Plus Add People is an AI image-to-image model built on Qwen’s Image-Edit-2509 framework. It is designed to generate and improve scenes featuring multiple individuals, to showcase them interacting naturally within a single setting. This model truly shines in grasping spatial relationships and crafting believable group dynamics, whether it’s a cozy family gathering, friends enjoying TV time, or crowds in diverse environments.
Qwen Image Edit Plus Add People also offers flexible multi-image blending via LoRA model support, making it a powerful tool for composing realistic multi-person scenes without complex manual edits, and an asset for digital artists, game developers, and content creators.
Key Features of Qwen Image Edit Plus Add People
- Multi-character scene synthesis with natural positioning and interaction dynamics.
- Multiple image input support for blending up to three source images seamlessly.
- LoRA model integration including pre-configured "add_people" for character insertion.
- Flexible aspect ratios from square 1:1 to cinematic 21:9, plus match-input-image option.
- Seed-based reproducibility for consistent results across iterations.
- High-quality output options with configurable format (JPEG, PNG, WebP) and quality control.
- Angle and perspective handling for diverse character viewpoints within scenes.
Best Use Cases
-
Marketing and Advertising: It can be utilized to create lifestyle imagery with multiple models in realistic settings without expensive photoshoots.
-
Game Development: It can generate concept art showing character groups, party interactions, or crowd scenes for world-building.
-
Social Media Content: It can precisely edit existing photos to add or modify people in scenes while maintaining visual coherence.
-
Film Pre-visualization: It is capable of creating mock-up multi-character scenes for storyboarding and blocking decisions.
-
E-commerce: It can effortlessly enhance product photography by adding contextual human elements showing usage scenarios.
-
Architectural Visualization: It can populate renders with people at various scales and angles to demonstrate space functionality.
Prompt Tips and Output Quality
-
Effective prompting is essential for precise visual outputs; therefore, prompts should be specific about quantity and activity. So, instead of "add people," use "two adults sitting on a couch watching television" or "family of four preparing dinner in the kitchen."
-
Multi-image workflow: Use 'image_1' as a primary scene, 'image_2' for character references or environmental elements to blend, and 'image_3' for additional atmospheric details. The model intelligently composites these inputs based on the prompt guidance.
-
LoRA selection matters: The 'add_people' LoRA is optimized for character insertion; therefore, for specialized effects, experiment with custom LoRA URLs in the advanced parameters.
-
Aspect ratio strategy: Use the aspect ratio to match your intended platform: '1:1' for social media posts, '16:9' for YouTube thumbnails, '9:16' for vertical video formats. Additionally, you can use 'match_input_image' when preserving original dimensions is essential to the visual output.
-
Seed control: Set a specific seed (not -1) when iterating on a composition to maintain consistent character placement while refining prompt details.
-
Quality settings: For final outputs, use the 'PNG' format at '85+' to preserve edge details crucial for multi-character scenes; you may lower the settings for drafts.
FAQs
Is Multiple Characters open-source?
The model is fine-tuned from Qwen/Qwen-Image-Edit-2509 and available through Segmind's platform. You may check the base model's license for specific usage terms.
How is it different from standard image-to-image models?
This model specializes in multi-character spatial coherence and interaction realism; thus, it understands group dynamics rather than treating each person as an isolated element, making it superior to generic editing models.
What's the maximum number of characters I can add?
The model can handle multiple characters effectively, and the optimal results can be achieved with 2-6 distinct people (objects). Larger crowds may require iterative approaches or multiple generations combined in post-processing.
Can I control specific character positions?
Yes, through detailed prompts that clearly specify locations (e.g., "person on left side facing right, second person in center background"; combining this with seed control enables precise positioning.
What parameters should I tweak for the best results?
Start with 'aspect_ratio' that matches your use case, set a consistent seed for iterations, and use the 'add_people' LoRA.
For production work, prioritize PNG format with the '85-95' quality.
Does it work with existing photos or only the generated images?
It can work efficiently with existing photos and the generated images. You can upload your photos as 'image_1' and use prompts to add, remove, or modify people within the existing scene context.
Popular Models
