# Flux Lora | Portrait Trainer

Optimized FLUX LoRA training for portrait generation with vivid highlights and highly detailed results.

## API Information

- **Model Slug:** flux-lora-portrait-trainer
- **Branded URL:** https://www.eachlabs.ai/black-forest-labs/flux-lora/flux-lora-portrait-trainer
- **Provider:** Black Forest Labs
- **Category:** Training
- **Output Type:** code
- **Status:** active
- **Version:** 0.0.1
- **Base Cost:** Within 1000-step minimum: flat $2.40
- **Estimated Processing Time:** 225 seconds
- **Last Updated:** 2026-06-10
- **Interactive Demo:** https://www.eachlabs.ai/ai-models/flux-lora-portrait-trainer

## Pricing

- **Charge Type:** dynamic
- **Estimated Price (default example):** $2.40
- **Pricing Details:** Within 1000-step minimum: flat $2.40

### Pricing Rules

| Condition | Pricing |
| --- | --- |
| steps <= "1000" | Within 1000-step minimum: flat $2.40 |
| steps > "1000" | Over 1000 steps: $0.0024 per step on the whole step count |
| Default (fallback) | $2.40 |

## Input Schema

| Parameter | Type | Required | Default | Constraints | Description |
|-----------|------|----------|---------|-------------|-------------|
| images_data_url | string | Yes | - | zip | Only a ZIP archive URL is accepted. The ZIP must contain at least 10 images of a consistent style (more is recommended). Optionally, you can include text files (captions) with the same name as each image. If the captions contain the string [trigger] and a trigger_word parameter is provided, [trigger] will automatically be replaced with that word. |
| trigger_phrase | string | No | - | - | Trigger phrase to be used in the captions. If None, a trigger word will not be used.  If no captions are provide the trigger_work will be used instead of captions. If captions are provided, the trigger word will replace the `[trigger]` string in the captions. |
| steps | integer | No | 1000 | 1–10000 | Number of steps to train the LoRA on. |
| multiresolution_training | boolean | No | true | - | If True, multiresolution training will be used. |
| subject_crop | boolean | No | true | - | If True, the subject will be cropped from the image. |
| data_archive_format | string | No | - | - | The format of the archive. If not specified, the format will be inferred from the URL. |
| resume_from_checkpoint | string | No | - | - | URL to a checkpoint to resume training from. |
| create_masks | boolean | No | false | - | If True, masks will be created for the subject. |
| learning_rate | number | No | 0.00009 | 0.000001–0.001 | Learning rate to use for training. |

## Example Request

```bash
curl -X POST https://api.eachlabs.ai/v1/prediction/ \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
  "model": "flux-lora-portrait-trainer",
  "input": {
    "images_data_url": "https://your-asset-url"
  }
}'
```

## Output Schema

Response returned by `GET /v1/prediction/{id}` when the job completes:

```json
{
  "status": "success",
  "predictionID": "string",
  "output": "code",
  "metrics": {
    "predict_time": "number (seconds)"
  }
}
```

## Polling

```bash
curl https://api.eachlabs.ai/v1/prediction/{PREDICTION_ID} \
  -H "Authorization: Bearer YOUR_API_KEY"
```

| Status | Meaning |
|--------|---------|
| `processing` | Still running — poll again |
| `success` | Done — read `output` |
| `error` | Failed — read `message` / `details` |

## Webhook (alternative to polling)

Pass `"webhook_url": "https://your.host/path"` in the create request. Eachlabs POSTs this payload when the job ends:

```json
{
  "exec_id": "prediction-uuid",
  "status": "succeeded",
  "output": "https://...",
  "error": ""
}
```

`status` is `"succeeded"` or `"failed"`. `exec_id` equals the `predictionID` from create. Return 2xx within 30 seconds.

## Errors

Error body: `{ "status": "error", "message": "...", "details": "..." }`

| Code | Meaning |
|------|---------|
| `400` | Invalid input |
| `401` | Missing / invalid `Authorization` bearer token |
| `404` | Unknown model or prediction id |
| `429` | Rate limit — 100 creates / min, 10 concurrent per key |
| `5xx` | Retry with backoff |

## Overview

**flux-lora-portrait-trainer — Training AI Model**

flux-lora-portrait-trainer from Black Forest Labs empowers developers and creators to fine-tune FLUX models specifically for generating stunning portraits with vivid highlights and intricate details. Part of the flux-lora family, this training tool optimizes LoRA adapters on the high-performance FLUX.2 architecture, enabling custom portrait generation that excels in photorealistic facial rendering and expressive lighting. Ideal for those searching for **flux-lora-portrait-trainer API** or **Black Forest Labs training** solutions, it delivers highly detailed results up to 4MP resolution with any aspect ratio.

Unlike generic training scripts, flux-lora-portrait-trainer leverages FLUX.2's rectified flow transformer design, distilled for efficiency while preserving quality for portrait-specific adaptations. This makes it perfect for building specialized AI models that capture nuanced skin tones, dynamic poses, and professional lighting in portraits.

## Usage Notes

- API Base URL: `https://api.eachlabs.ai/v1`
- Authentication: send `Authorization: Bearer YOUR_API_KEY`. Generate a key from the Eachlabs dashboard at https://www.eachlabs.ai/dashboard/api-keys.
- File-typed parameters (`*_url`, `image_url`, `video_url`, `audio_url`, etc.) accept publicly-reachable HTTPS URLs only. Upload your asset first (GCS / S3 / your CDN) and pass the resulting URL. Data-URIs and localhost URLs are rejected.
- For structured parameters (arrays / objects) send real JSON values, not stringified payloads.
- Monetary values are reported in USD; per-token / per-megapixel rates may be billed in micro-cents internally.
- Prefer `webhook_url` over polling for long-running predictions — see the Webhook Callback section.