Higgsfield Soul Image To Image

Higgsfield Soul Image To Image

Playground

SOUL is a realistic image-to-image engine for sophisticated visuals that, with Soul ID, preserves character consistency across scenes. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.

Features

Higgsfield SOUL Image-to-Image

Higgsfield SOUL Image-to-Image is a high-quality image transformation model built for realism, tasteful aesthetics, and strong character consistency. It is designed for reference-driven image editing workflows where you want to restyle, reframe, or evolve a subject while preserving identity, structure, and visual coherence.

Why Choose This?

• Elegant realism Produce believable lighting, natural skin and hair, and clean material rendering without an over-processed look.

• Strong character consistency Keep the same identity across different scenes, poses, outfits, and lighting conditions.

• Faithful image-to-image editing Preserve core composition and structure while changing styling, color, mood, wardrobe, or background.

• Art-direction friendly Respond well to cinematic language such as lens cues, depth of field, rim light, palette direction, and grading intent.

• Wide stylistic range Move between photoreal, editorial, painterly, or graphic looks without heavy prompt engineering.

Parameters

How to Use

1. Upload a reference image — use a clean, well-lit image with minimal occlusion.
2. Add a Soul ID (optional) — reuse the same ID if you want stronger identity continuity across multiple shots.
3. Write your prompt — describe the subject, setting, outfit, props, lighting, lens feel, palette, and mood.
4. Choose quality — use medium for lower cost or high for stronger final quality.
5. Set aspect ratio — choose the composition format that fits your delivery needs.
6. Adjust strength — use lower values for source-faithful edits and higher values for looser restyling.
7. Submit — generate the image and iterate as needed.

Example Prompt

Turn this portrait into a cinematic editorial fashion shot at golden hour, soft rim light, clean skin texture, natural color grading, luxury wardrobe styling, realistic background depth

Pricing

Pricing depends on the selected quality tier.

Billing Rules

• medium costs $0.09 per image
• high costs $0.19 per image
• Pricing depends only on the selected quality tier

Best Use Cases

• Character lookbooks — Maintain one identity across many styled outputs.
• Narrative keyframes — Generate consistent visual storytelling frames from a reference.
• Brand mascots and recurring talent — Keep a recognizable subject across multiple campaigns.
• Fashion and editorial sets — Explore wardrobe, lighting, and scene variations with consistency.
• Advertising creatives — Produce campaign assets with stable character identity.
• Storyboards and pre-visualization — Quickly develop coherent visual directions for production planning.

Pro Tips

• Use a clean, front-facing, well-lit reference when identity preservation matters.
• Keep prompts specific about what should change and what should remain consistent.
• Reuse the same soul_id when building a sequence or visual set around the same subject.
• Use lower strength when structure fidelity matters more than style variation.
• Use high quality for final selects and medium when exploring directions more quickly.

Notes

• This model is designed for image-to-image workflows with strong identity and style consistency.
• Reusing the same soul_id can improve continuity across generations.
• Better source images generally produce more stable and higher-quality results.
• Pricing is fixed by quality tier.

Related Models

• Other Higgsfield image generation workflows — Useful when you need text-to-image generation instead of reference-based editing.
• Character-consistent generation workflows — Useful when maintaining identity across multiple scenes is the main priority.

# 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}"