1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
const axios = require('axios');
const api_key = "YOUR API-KEY";
const url = "https://api.segmind.com/v1/higgsfield-soul-2";
const data = {
"prompt": "Editorial fashion portrait of a woman with auburn wavy hair and freckles, cinematic golden-hour window light, soft film grain, high-end magazine aesthetic, cream cashmere sweater, shallow depth of field, shot on 85mm",
"aspect_ratio": "3:4",
"resolution": "1080p",
"image_url": "https://segmind-resources.s3.amazonaws.com/input/higgsfield-soul-2-input-reference.jpg",
"seed": 123456
};
(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 describing your image; required. Use editorial, fashion, camera and lighting terms.
Output frame ratio; required enum. Use 3:4 portraits, 9:16 social, 16:9 wide.
Allowed values:
Output detail; required, 720p or 1080p. Pick 1080p for campaigns, 720p for drafts.
Allowed values:
Optional reference image URL guiding style and composition. Supply for moodboard or campaign matching.
Optional integer, 1 to 1000000, for reproducibility. Fix to repeat a result.
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.
Higgsfield Soul 2.0 — Fashion-Aware AI Image Generation
What is Higgsfield Soul 2.0?
Higgsfield Soul 2.0 is Higgsfield's next-generation foundation photo model, built for creative, fashion-aware, culture-native image generation with aesthetic taste baked into every output. It succeeds the original Soul model and expands its hyper-realistic, smartphone-style roots into high-fashion and editorial territory. Rather than producing generic renders, Soul 2.0 generates images that feel photographed rather than synthesized — art-directed compositions with intentional styling, lighting, and mood you would expect from a magazine spread.
On Segmind, you call Soul 2.0 through a single synchronous API. Send a prompt, choose an aspect_ratio and resolution, and optionally pass an image_url reference and a seed. The model returns a finished image directly in the response — no polling required.
Key Features
- Fashion and culture fluency — Soul 2.0 understands fashion terms, camera language, and internet aesthetics like Y2K, coquette, Frutiger Aero, and street style, rendering them accurately instead of as costume-party approximations.
- Editorial realism — Skin texture, fabric detail, natural lighting, and clean compositions built by a team with luxury fashion and brand backgrounds, reducing the plastic look common in generic models.
- Reference-guided generation — Pass an image URL to transfer vibe, composition, palette, or era into a fresh generation.
- Flexible framing — Aspect ratios span 1:1, 4:3, 3:4, 16:9, and 9:16, with 720p and 1080p resolution options.
- Reproducibility — A seed range of 1 to 1,000,000 locks results for repeatable iteration.
Best Use Cases
Soul 2.0 excels at fashion editorials, campaign visuals, lookbooks, and e-commerce photography. Creators use it for UGC and lifestyle content at scale, social ad creatives in bulk, album and artist visuals, and virtual influencer or avatar imagery. Because taste is built in, short prompts paired with strong style cues produce magazine-ready portraits without heavy prompt engineering. It is a natural fit for brand teams that need cohesive, production-ready visuals fast.
Prompt Tips and Output Quality
Prompt like a photographer: name the medium (iPhone, film, disposable, Polaroid, Kodak Portra), the lens (85mm), the light, and the styling. Soul 2.0 adjusts grain, color science, and dynamic range accordingly. Keep prompts focused — under roughly 100 words — and let the model's built-in aesthetic handle the rest. For a specific look, supply a reference image. Note that fine text on garments and props can render imperfectly, so review close-up details before final use.
FAQs
Is Higgsfield Soul 2.0 a text-to-image or image-to-image model? Both. Generate purely from a text prompt, or supply an image URL as a style and composition reference to guide the output.
What aspect ratios and resolutions does it support? Aspect ratios are 1:1, 4:3, 3:4, 16:9, and 9:16. Resolution options are 720p and 1080p; 3:4 at 1080p is a strong default for portraits.
How is Soul 2.0 different from the original Soul? Soul 2.0 keeps the realistic, smartphone-style foundation but extends it into high-fashion and editorial territory with stronger culture and camera awareness.
How does it compare to Midjourney or Seedream? Soul 2.0 is tuned specifically for fashion realism and creative direction, while alternatives like Seedream and Midjourney target broader general-purpose generation and editing.
Can I reproduce the same image again?
Yes. Set the optional seed (1 to 1,000,000) to keep results repeatable across runs while you refine a prompt.
Popular Models
