Higgsfield Soul Image To Image
Playground
Try it on WavespeedAI!SOUL is the foundational engine of the Higgsfield visual suite, a model that renders images with realism and sophisticated aesthetics. Its true power is unlocked when paired with our Soul ID technology, which allows you to maintain perfect character consistency across countless new scenes and styles.
Features
Higgsfield SOUL — Image-to-Image ✨
SOUL is the foundational engine of Higgsfield’s visual suite—an image model tuned for realism and tasteful aesthetics. It delivers rock-solid character consistency across new scenes, poses, outfits, and lighting—turn one reference into a whole visual universe. 🌌
Why it shines 🌟
- Elegant realism — Believable lighting, natural skin/hair, clean materials—no plastic over-sharpening. 🎨
- ID consistency — Lock a character’s identity and regenerate them across countless settings and styles. 🪪🔁
- Faithful I2I edits — Preserve structure while restyling color, mood, wardrobe, and background. 🧥🌆
- Art-direction friendly — Responds to camera verbs, palette cues, and film language (DOF, rim light, grade). 🎬
- Wide style range — From photoreal and editorial to painterly and graphic—switch looks without prompt gymnastics.
Core capabilities 🧰
- Image-to-Image (I2I) — Restyle a reference while keeping geometry intact. 🪄
- Character persistence — Reuse an ID to retain facial features and identity. 👤✅
- Style transfer — Carry palettes, textures, and lighting into new compositions. 🎨
- In-image text guidance — Titles, labels, and graphic elements with layout awareness. 🔖
How to use 🚀
- Provide a reference image — clean, well-lit, minimal occlusions. 🖼️
- (Optional) Attach a Soul ID — keep the same ID to maintain identity across shots. 🪪
- Write your prompt — subject, setting, wardrobe/props, lighting, lens/POV, palette, mood. ✍️
- Choose a style — subtle restyle vs. bold aesthetic shift. ⚖️
- Set size/aspect — pick feed (4:5), story (9:16), or widescreen (16:9). 📐
- Set strength — lower = closer to source; higher = freer restyle. 🎛️
- Generate → iterate — tweak style, lighting, or camera language; keep the same Soul ID for continuity. 🔄
Price 💵
Quality | Price per Image |
---|---|
Medium | $0.025 |
High | $0.030 |
Common use cases 📝
- Character lookbooks
- Narrative keyframes
- Brand mascots across seasons
- Fashion/editorial sets
- Ad creatives with consistent talent
- Storyboards & pre-viz
Authentication
For authentication details, please refer to the Authentication Guide.
API Endpoints
Submit Task & Query Result
# Submit the task
curl --location --request POST "https://api.wavespeed.ai/api/v3/higgsfield/soul/image-to-image" \
--header "Content-Type: application/json" \
--header "Authorization: Bearer ${WAVESPEED_API_KEY}" \
--data-raw '{
"size": "1152*2048",
"style": "Creatures",
"strength": 1,
"quality": "medium"
}'
# Get the result
curl --location --request GET "https://api.wavespeed.ai/api/v3/predictions/${requestId}/result" \
--header "Authorization: Bearer ${WAVESPEED_API_KEY}"
Parameters
Task Submission Parameters
Request Parameters
Parameter | Type | Required | Default | Range | Description |
---|---|---|---|---|---|
prompt | string | Yes | - | The positive prompt for the generation. | |
image | string | Yes | - | ||
size | string | Yes | 1152*2048 | 960*1696, 1088*1632, 1152*1536, 1152*2048, 1536*1536, 1536*1152, 1536*2048, 1344*2016, 1632*1088, 1696*960, 2016*1344, 2048*1152, 2048*1536 | The size of the generated media in pixels (width*height). |
style | string | No | Creatures | Creatures, Medieval, Spotlight, Giant People, Red balloon, green editorial, Subway, Library, Realistic, DigitalCam, Grillz Selfie, Bleached Brows, Sitting on the Street, Crossing the street, Angel Wings, Duplicate, Quiet luxury, Fireproof, Elevator Mirror, 360 cam, Glitch, FashionShow, PixeletedFace, Sunbathing, Paper Face, 90s Grain, Geominimal, Foggy Morning, Overexposed, Sunset beach, Giant Accessory, RingSelfie, Street view, 90’s Editorial, Rhyme & blues, 2000s Cam, CCTV, 0.5 Outfit, Amalfi Summer, Bimbocore, 0.5 Selfie, Sand, Vintage PhotoBooth, afterparty cam, Babydoll MakeUp, Through The Glass, Gallery, Eating Food, Swords Hill, Office beach, Help It's Too Big, Japandi, iPhone, Gorpcore, Indie sleaze, Fairycore, Tumblr, Avant-garde, HairClips, birthday mess, Clouded Dream, Y2K Posters, tokyo drift, Object Makeup, Graffiti, Sunburnt, hallway noir, 2000s Fashion, Night Beach, Movie, Long legs, 7\, General, Nail Check, Coquette core, Mixed Media, Selfcare, Grunge, Double take, 505room, Flight mode, Escalator, burgundy suit, Fisheye, Shoe Check, Rainy Day, Mt. Fuji, Sea breeze, Invertethereal, Y2K, Tokyo Streetstyle, chrome exit, Night rider, Artwork, Glazed doll skin makeup, mount view, 2049, blackout fit, Bike mafia, static glow, Nicotine glow, brick shade, dmv, Fish-eye twin, It’s french, cocktail | Chosen preset for soul image generation |
strength | number | No | 1 | 0.00 ~ 1.00 | The strength to use for the style. |
quality | string | No | medium | medium, high | The resolution of the output image. |
seed | integer | No | - | -1 ~ 2147483647 | The random seed to use for the generation. -1 means a random seed will be used. |
Response Parameters
Parameter | Type | Description |
---|---|---|
code | integer | HTTP status code (e.g., 200 for success) |
message | string | Status message (e.g., “success”) |
data.id | string | Unique identifier for the prediction, Task Id |
data.model | string | Model ID used for the prediction |
data.outputs | array | Array of URLs to the generated content (empty when status is not completed ) |
data.urls | object | Object containing related API endpoints |
data.urls.get | string | URL to retrieve the prediction result |
data.has_nsfw_contents | array | Array of boolean values indicating NSFW detection for each output |
data.status | string | Status of the task: created , processing , completed , or failed |
data.created_at | string | ISO timestamp of when the request was created (e.g., “2023-04-01T12:34:56.789Z”) |
data.error | string | Error message (empty if no error occurred) |
data.timings | object | Object containing timing details |
data.timings.inference | integer | Inference time in milliseconds |