# MOSS Image LoRA Trainer

> Train custom MOSS Image LoRA models efficiently with flexible batch sizes, optimized attention, and 12–24 GB VRAM support for high-quality 512×512 to 1024×1024 image generation.

## Overview

- **Model ID**: `moss-image-lora-trainer`
- **Category**: training
- **Provider**: modelslab
- **Source Type**: Open Source Model
- **Status**: active
- **Screenshot**: `https://assets.modelslab.ai/generations/7cc607c2-1e9f-454b-b6a8-f552e2b1924b.png`

## API Information

This model can be used via our HTTP API. See the API documentation and usage examples below.

### Endpoint

- **URL**: `https://modelslab.com/api/v6/trainer/train`
- **Method**: POST

### Parameters

- **`instance_prompt`** (required): Prompt for training the instance.
  - Type: textarea

- **`images`** (required): Dataset for training, either as a path string or a list of paths (Required)
  - Type: array

- **`hf_token`** (optional): Hugging Face token for accessing models.
  - Type: text

- **`hf_username`** (optional): Hugging Face username for model storage
  - Type: text

- **`base_model_type`** (optional): Type of the base model. (Required)
  - Type: text

- **`seed`** (optional): Random seed for reproducibility. (Required)
  - Type: number

- **`training_steps`** (optional): 
  - Type: number

- **`resolution`** (optional): Resolution for the training images.
  - Type: number (range: 512-1024)

- **`batch_size`** (optional): Batch size for training.
  - Type: number (range: 1-128)

- **`mixed_precision`** (optional): Precision mode for training allowed (fp16, bf16”)
  - Type: select (options: fp16)

- **`rank`** (optional): Rank for LoRA
  - Type: number (range: 1-512)

- **`alpha`** (optional): Alpha for LoRA.
  - Type: number (range: 1-1024)

- **`learning_rate`** (optional): 
  - Type: number

- **`trainer_id`** (optional): 
  - Type: text

## Usage Examples

### cURL

```bash
curl --request POST \
  --url https://modelslab.com/api/v6/trainer/train \
  --header "Content-Type: application/json" \
  --data '{
    "key": "YOUR_API_KEY",
    "model_id": "moss-image-lora-trainer",
    "instance_prompt": "thalman than",
    "base_model_type": "moss-image-lora-trainer",
    "seed": "298329",
    "training_steps": "1500",
    "resolution": "1024",
    "batch_size": "4",
    "mixed_precision": "fp16",
    "rank": "4",
    "alpha": "2",
    "learning_rate": "0.0001",
    "trainer_id": "moss-image-lora-trainer"
  }'
```

### Python

```python
import requests

response = requests.post(
    "https://modelslab.com/api/v6/trainer/train",
    headers={
        "Content-Type": "application/json"
    },
    json={
        "key": "YOUR_API_KEY",
        "model_id": "moss-image-lora-trainer",
        "instance_prompt": "thalman than",
        "base_model_type": "moss-image-lora-trainer",
        "seed": "298329",
        "training_steps": "1500",
        "resolution": "1024",
        "batch_size": "4",
        "mixed_precision": "fp16",
        "rank": "4",
        "alpha": "2",
        "learning_rate": "0.0001",
        "trainer_id": "moss-image-lora-trainer"
    }
)

print(response.json())
```

### JavaScript

```javascript
fetch("https://modelslab.com/api/v6/trainer/train", {
  method: "POST",
  headers: {
    "Content-Type": "application/json"
  },
  body: JSON.stringify({
    "key": "YOUR_API_KEY",
    "model_id": "moss-image-lora-trainer",
    "instance_prompt": "thalman than",
    "base_model_type": "moss-image-lora-trainer",
    "seed": "298329",
    "training_steps": "1500",
    "resolution": "1024",
    "batch_size": "4",
    "mixed_precision": "fp16",
    "rank": "4",
    "alpha": "2",
    "learning_rate": "0.0001",
    "trainer_id": "moss-image-lora-trainer"
  })
})
.then(response => response.json())
.then(data => console.log(data));
```

## Integration Options

### CLI

Install: `curl -fsSL https://modelslab.sh/install.sh | sh` or `brew install modelslab/tap/modelslab`

```bash
modelslab auth login                                    # Authenticate
modelslab models search "moss-image-lora-trainer"                       # Find this model
modelslab models detail --id moss-image-lora-trainer                    # Get model details
modelslab generate image --prompt "..." --model moss-image-lora-trainer  # Generate with model
modelslab config set generation.default_model moss-image-lora-trainer   # Set as default
```

- Website: https://modelslab.sh
- GitHub: https://github.com/ModelsLab/modelslab-cli

### MCP Servers (Model Context Protocol)

For Claude Code, Cursor, VS Code, Windsurf, and any MCP-compatible agent.

- Generation: `https://modelslab.com/mcp/v7` (API key auth, 23 tools)
- Agent Control Plane: `https://modelslab.com/mcp/agents` (Bearer token auth, 10 tools)

Claude Code config (`~/.claude/settings.json`):

```json
{
  "mcpServers": {
    "modelslab-v7": {
      "url": "https://modelslab.com/mcp/v7",
      "headers": {
        "Authorization": "Bearer YOUR_API_KEY"
      }
    }
  }
}
```

- Documentation: https://docs.modelslab.com/mcp-web-api/overview

### Agent Skills

Install skill files directly into AI coding agents:

```bash
npx skills add modelslab/skills --all                          # All skills
npx skills add modelslab/skills --skill image-generation       # Specific skill
npx skills add modelslab/skills --all -a claude-code -a cursor # Target agents
```

- GitHub: https://github.com/ModelsLab/skills
- Documentation: https://docs.modelslab.com/agent-skills

### SDKs

- Python: `pip install modelslab`
- TypeScript: `npm install modelslab`
- PHP: `composer require modelslab/modelslab`
- Go: `go get github.com/modelslab/modelslab-go`
- Dart: `dart pub add modelslab`

## Links

- [Model Playground](https://modelslab.com/models/modelslab/moss-image-lora-trainer)
- [API Documentation](https://docs.modelslab.com)
- [ModelsLab Platform](https://modelslab.com)