skills.rest
kekoesports
kekoesportsCOMMUNITY
Explore all skills by this author →
Website𝕏@kekOesportsGitHub

higgsfield-soul-id

Train a personalized Soul model from 5–20 photos for identity-faithful image and video generation.

Updated Mar 17, 2026
Search Tags:#image-generation#video-generation#digital-twin#higgsfield#soul-id#face-training#identity-model
One-click install
npx skills add https://github.com/kekoesports/proyectozack --skill higgsfield-soul-id-kekoesports
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: higgsfield-soul-id
Source: https://github.com/kekoesports/proyectozack/tree/main/.agents/skills/higgsfield-soul-id
Command: npx skills add https://github.com/kekoesports/proyectozack --skill higgsfield-soul-id-kekoesports

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Train a Soul Character — a personalized model for identity-faithful image and video generation.

Core Features & Use Cases

  • Bootstrap and train a Soul from 5–20 photos of a single person with guidance for quality and variety.
  • Use the trained model with higgsfield generate to produce image or cinematic outputs via --soul-id <id>.
  • Documentation and troubleshooting references are available in the references/ directory.

Quick Start

Provide a name and 5–20 photos to train a Soul, then wait for the generated reference id and use it with higgsfield generate.

Frequently Asked Questions about higgsfield-soul-id

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I train a digital twin model for identity-faithful image generation?▼

After training your personal Soul model, you execute the `higgsfield generate` command with the `--soul-id` flag. This applies your trained identity model to produce personalized image or cinematic video outputs.

What photo requirements are needed to train a personal identity model?▼

Training a personal identity model requires 5 to 20 photos of a single person. The reference documentation provides specific guidance on photo quality and variety needed to accurately capture facial features.

How do I use a trained Soul ID with higgsfield generate for video outputs?▼

After training your personal Soul model, you execute the higgsfield generate command with the --soul-id flag. This applies your trained identity model to produce personalized image or cinematic video outputs.

Do I need the higgsfield CLI to create and use a Soul model?▼

Yes, the higgsfield CLI is required to create and use a Soul model. The CLI processes your input photos for training and handles the downstream identity-faithful image and cinematic generation commands.

Can I train a Soul model using photos of multiple people at once?▼

No, you cannot train a Soul model using photos of multiple people at once. The training process is designed for single-person identity capture, requiring 5 to 20 photos of one individual to ensure identity faithfulness.

Related Skills

Explore More Design & Creative Skills→
agentik-osagentik-os
10

higgsfield-soul-id

Train a personalized Soul avatar from 5–20 user photos for identity-consistent visuals.

Community
Intermediate
1veso1veso

higgsfield-soul-id

Train a personalized face-faithful identity model from photos using Bash scripts.

Community
Intermediate
ag47-ptag47-pt
1

higgsfield-soul-id

Train a personalized Higgsfield Soul identity model from face photos.

Community
Intermediate
ag47-ptag47-pt
1

google-labs-soul-id

Train Imagen 3 subject models for identity-consistent image generation.

Community
Intermediate