How do I install the Flux Lora Portrait Trainer skill?

Run npx skillmds@latest add awesome-genmedia/flux-lora-portrait-trainer in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Flux Lora Portrait Trainer safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Flux Lora Portrait Trainer?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Flux Lora Portrait Trainer free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Flux Lora Portrait Trainer?

awesome-genmedia (@awesome-genmedia) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Flux Lora Portrait Trainer by awesome-genmedia

Name: Flux Lora Portrait Trainer
Author: awesome-genmedia

Flux Lora Portrait Trainer

Train a custom LoRA model from portrait images. Trigger: Use when the user wants to train a custom LoRA, or requests 'train lora', 'flux portrait trainer', 'custom lora training', or 'train model on my photos'.

File contents

Flux Lora | Portrait Trainer

Train a custom LoRA model from a set of portrait images. Upload a ZIP archive of images and configure training parameters to create a personalized LoRA weight file.

Quick Start

curl -X POST https://api.eachlabs.ai/v1/prediction \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $EACHLABS_API_KEY" \
  -d '{
    "model": "flux-lora-portrait-trainer",
    "version": "0.0.1",
    "input": {
      "images_data_url": "https://example.com/portrait-images.zip",
      "trigger_phrase": "photo of sks person",
      "steps": 1000,
      "learning_rate": 0.00009,
      "create_masks": true,
      "subject_crop": true
    }
  }'

Parameters

Parameter	Type	Default	Description
`create_masks`	boolean	`false`	If True, masks will be created for the subject
`data_archive_format`	string	(empty)	The format of the archive. If not specified, inferred from the file
`images_data_url`	string	(empty)	ZIP archive URL containing at least 10 images of a subject
`learning_rate`	number	`0.00009`	Learning rate for training
`multiresolution_training`	boolean	`true`	If True, multiresolution training will be used
`resume_from_checkpoint`	string	(empty)	URL to a checkpoint to resume training from
`steps`	integer	`1000`	Number of training steps
`subject_crop`	boolean	`true`	If True, the subject will be cropped from the image
`trigger_phrase`	string	(empty)	Trigger phrase to use in captions. If None, no trigger word is used

Output

Type: code

Examples

Basic Portrait LoRA Training

{
  "model": "flux-lora-portrait-trainer",
  "version": "0.0.1",
  "input": {
    "images_data_url": "https://example.com/my-portraits.zip",
    "trigger_phrase": "photo of johndoe person",
    "steps": 1200,
    "learning_rate": 0.00009,
    "create_masks": true,
    "subject_crop": true,
    "multiresolution_training": true
  }
}

Resume Training from Checkpoint

{
  "model": "flux-lora-portrait-trainer",
  "version": "0.0.1",
  "input": {
    "images_data_url": "https://example.com/more-portraits.zip",
    "trigger_phrase": "photo of johndoe person",
    "steps": 500,
    "resume_from_checkpoint": "https://example.com/checkpoint-1000.safetensors"
  }
}

Related Models

Juggernaut Flux Lora - Use trained LoRA for image generation
Flux Kontext Lora Text to Image - Text-to-image with LoRA
Flux Krea Lora Image to Image - Image-to-image with LoRA

Documentation

EachLabs API Docs

---name: flux-lora-portrait-trainerdescription: Train a custom LoRA model from portrait images. Trigger: Use when the user wants to train a custom LoRA, or requests 'train lora', 'flux portrait trainer', 'custom lora training', or 'train model on my photos'.---# Flux Lora | Portrait TrainerTrain a custom LoRA model from a set of portrait images. Upload a ZIP archive of images and configure training parameters to create a personalized LoRA weight file.## Quick Start```bashcurl -X POST https://api.eachlabs.ai/v1/prediction \ -H "Content-Type: application/json" \ -H "X-API-Key: $EACHLABS_API_KEY" \ -d '{ "model": "flux-lora-portrait-trainer", "version": "0.0.1", "input": { "images_data_url": "https://example.com/portrait-images.zip", "trigger_phrase": "photo of sks person", "steps": 1000, "learning_rate": 0.00009, "create_masks": true, "subject_crop": true } }'```## Parameters| Parameter | Type | Default | Description ||-----------|------|---------|-------------|| `create_masks` | boolean | `false` | If True, masks will be created for the subject || `data_archive_format` | string | (empty) | The format of the archive. If not specified, inferred from the file || `images_data_url` | string | (empty) | ZIP archive URL containing at least 10 images of a subject || `learning_rate` | number | `0.00009` | Learning rate for training || `multiresolution_training` | boolean | `true` | If True, multiresolution training will be used || `resume_from_checkpoint` | string | (empty) | URL to a checkpoint to resume training from || `steps` | integer | `1000` | Number of training steps || `subject_crop` | boolean | `true` | If True, the subject will be cropped from the image || `trigger_phrase` | string | (empty) | Trigger phrase to use in captions. If None, no trigger word is used |## Output- **Type:** code## Examples### Basic Portrait LoRA Training```json{ "model": "flux-lora-portrait-trainer", "version": "0.0.1", "input": { "images_data_url": "https://example.com/my-portraits.zip", "trigger_phrase": "photo of johndoe person", "steps": 1200, "learning_rate": 0.00009, "create_masks": true, "subject_crop": true, "multiresolution_training": true }}```### Resume Training from Checkpoint```json{ "model": "flux-lora-portrait-trainer", "version": "0.0.1", "input": { "images_data_url": "https://example.com/more-portraits.zip", "trigger_phrase": "photo of johndoe person", "steps": 500, "resume_from_checkpoint": "https://example.com/checkpoint-1000.safetensors" }}```## Related Models- [Juggernaut Flux Lora](../juggernaut-flux-lora/SKILL.md) - Use trained LoRA for image generation- [Flux Kontext Lora Text to Image](../flux-kontext-lora-text-to-image/SKILL.md) - Text-to-image with LoRA- [Flux Krea Lora Image to Image](../flux-krea-lora-image-to-image/SKILL.md) - Image-to-image with LoRA## Documentation- [EachLabs API Docs](https://docs.eachlabs.ai)

Frequently asked questions