# Moss Image Sft Text to Image

> 🎨 MOSS-Image-SFT transforms your text prompts into stunning, high-quality images with rich details, creative styles, and intelligent layout control. Bring your imagination to life with AI! ✨

## Overview

- **Model ID**: `moss-image-sft`
- **Category**: image
- **Provider**: modelslab
- **Source Type**: Open Source Model
- **Status**: model_ready
- **Screenshot**: `https://assets.modelslab.ai/generations/7e3c5523-f39d-4b8d-8561-8b1df5b1ee7b.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/images/text2img`
- **Method**: POST

### Parameters

- **`prompt`** (required): Enter the text for the image you want to generate
  - Type: textarea

- **`width`** (required): Width must be at least 512 px and at most 2048 px
  - Type: number (range: 512-2048)

- **`height`** (required): Height must be at least 512 px and at most 2048 px
  - Type: number (range: 512-2048)

- **`samples`** (required): Number of images that can be generated in a single request
  - Type: number (range: 1-2)

- **`num_inference_steps`** (optional): The number of denoising steps. Values range from 30 to 50, and any value above 50 will be capped at 50.
  - Type: number (range: 30-50)

## Usage Examples

### cURL

```bash
curl --request POST \
  --url https://modelslab.com/api/v6/images/text2img \
  --header "Content-Type: application/json" \
  --data '{
    "key": "YOUR_API_KEY",
    "model_id": "moss-image-sft",
    "prompt": "EDIT THE PROVIDED PHOTO — DO NOT CHANGE THE HUMAN SUBJECT.Create a premium editorial mixed-media photograph using the provided image as the base.HUMAN SUBJECT — ABSOLUTE LOCK:- Keep the exact original person from the reference photo.- Do NOT regenerate, redraw, beautify, reshape, retouch, or replace the person.- Preserve the exact face, facial structure, facial proportions, skin texture, hairstyle, cap, clothing, body shape, pose, posture, hand position, and expression.- Preserve the original smile and the exact candid “looking back” expression.- The human must remain photorealistic and visually identical to the source photograph.- No AI beauty filter, no face alteration, no skin smoothing, no facial reconstruction.- The person is the photographic anchor of the composition.ENVIRONMENT — TRANSFORM INTO EDITORIAL DOODLE COLLAGE:Transform ONLY the surrounding environment into a sophisticated hand-drawn editorial collage.Keep the original garden composition recognizable:- agave plants- tall columnar cacti- trees- sky- garden path- surrounding foliageAdd expressive hand-drawn doodles and graphic interventions around the environment:- rough black marker outlines- cream/white chalk strokes- mustard-yellow hand-drawn lines- arrows- stars- crowns- circles- underlines- sketch marks- imperfect handwritten annotations- playful graphic symbols- subtle torn-paper textures- distressed ink textures- analog scrapbook elements- subtle grain and print imperfectionsThe doodles should interact with the plants and background naturally, almost as if an editorial photographer photographed a real location and then an artist drew directly over the photograph.IMPORTANT:The doodles must NOT cover, distort, redraw, or modify the human subject.Keep a clean visual separation around the person's face, body, clothing, and silhouette.TYPOGRAPHY:Add sparse, tasteful handwritten editorial notes integrated into the environment, such as:“GOOD DAY”“SAME PLACE / DIFFERENT MOOD”“JUST A BOY”“MORE LIFE HERE”“NATURE ALWAYS A GOOD IDEA”Use imperfect handwritten lettering rather than polished digital typography.ART DIRECTION:Youthful contemporary fashion-editorial aesthetic.Analog scrapbook + street editorial + botanical photography.Energetic but sophisticated.Controlled visual chaos.Premium Instagram campaign aesthetic.Natural photographic lighting.Strong depth and foreground layering.Organic composition rather than a symmetrical graphic layout.COLOR & FINISH:Keep the original natural photographic colors of the human and environment.Use black, warm cream, muted mustard yellow, and subtle off-white for the doodle elements.Add subtle film grain, paper texture, and analog print character.Avoid excessive saturation.FINAL RESULT:A highly photogenic editorial photograph that looks like a real candid outdoor fashion/lifestyle photograph enhanced with hand-drawn graphic artwork.MOST IMPORTANT RULE:THE HUMAN SUBJECT MUST REMAIN EXACTLY AS IN THE ORIGINAL PHOTO.ONLY THE ENVIRONMENT AND GRAPHIC ELEMENTS MAY BE STYLIZED",
    "width": "1024",
    "height": "1024",
    "samples": "1",
    "num_inference_steps": "30"
  }'
```

### Python

```python
import requests

response = requests.post(
    "https://modelslab.com/api/v6/images/text2img",
    headers={
        "Content-Type": "application/json"
    },
    json={
        "key": "YOUR_API_KEY",
        "model_id": "moss-image-sft",
        "prompt": "EDIT THE PROVIDED PHOTO — DO NOT CHANGE THE HUMAN SUBJECT.Create a premium editorial mixed-media photograph using the provided image as the base.HUMAN SUBJECT — ABSOLUTE LOCK:- Keep the exact original person from the reference photo.- Do NOT regenerate, redraw, beautify, reshape, retouch, or replace the person.- Preserve the exact face, facial structure, facial proportions, skin texture, hairstyle, cap, clothing, body shape, pose, posture, hand position, and expression.- Preserve the original smile and the exact candid “looking back” expression.- The human must remain photorealistic and visually identical to the source photograph.- No AI beauty filter, no face alteration, no skin smoothing, no facial reconstruction.- The person is the photographic anchor of the composition.ENVIRONMENT — TRANSFORM INTO EDITORIAL DOODLE COLLAGE:Transform ONLY the surrounding environment into a sophisticated hand-drawn editorial collage.Keep the original garden composition recognizable:- agave plants- tall columnar cacti- trees- sky- garden path- surrounding foliageAdd expressive hand-drawn doodles and graphic interventions around the environment:- rough black marker outlines- cream/white chalk strokes- mustard-yellow hand-drawn lines- arrows- stars- crowns- circles- underlines- sketch marks- imperfect handwritten annotations- playful graphic symbols- subtle torn-paper textures- distressed ink textures- analog scrapbook elements- subtle grain and print imperfectionsThe doodles should interact with the plants and background naturally, almost as if an editorial photographer photographed a real location and then an artist drew directly over the photograph.IMPORTANT:The doodles must NOT cover, distort, redraw, or modify the human subject.Keep a clean visual separation around the person's face, body, clothing, and silhouette.TYPOGRAPHY:Add sparse, tasteful handwritten editorial notes integrated into the environment, such as:“GOOD DAY”“SAME PLACE / DIFFERENT MOOD”“JUST A BOY”“MORE LIFE HERE”“NATURE ALWAYS A GOOD IDEA”Use imperfect handwritten lettering rather than polished digital typography.ART DIRECTION:Youthful contemporary fashion-editorial aesthetic.Analog scrapbook + street editorial + botanical photography.Energetic but sophisticated.Controlled visual chaos.Premium Instagram campaign aesthetic.Natural photographic lighting.Strong depth and foreground layering.Organic composition rather than a symmetrical graphic layout.COLOR & FINISH:Keep the original natural photographic colors of the human and environment.Use black, warm cream, muted mustard yellow, and subtle off-white for the doodle elements.Add subtle film grain, paper texture, and analog print character.Avoid excessive saturation.FINAL RESULT:A highly photogenic editorial photograph that looks like a real candid outdoor fashion/lifestyle photograph enhanced with hand-drawn graphic artwork.MOST IMPORTANT RULE:THE HUMAN SUBJECT MUST REMAIN EXACTLY AS IN THE ORIGINAL PHOTO.ONLY THE ENVIRONMENT AND GRAPHIC ELEMENTS MAY BE STYLIZED",
        "width": "1024",
        "height": "1024",
        "samples": "1",
        "num_inference_steps": "30"
    }
)

print(response.json())
```

### JavaScript

```javascript
fetch("https://modelslab.com/api/v6/images/text2img", {
  method: "POST",
  headers: {
    "Content-Type": "application/json"
  },
  body: JSON.stringify({
    "key": "YOUR_API_KEY",
    "model_id": "moss-image-sft",
    "prompt": "EDIT THE PROVIDED PHOTO — DO NOT CHANGE THE HUMAN SUBJECT.Create a premium editorial mixed-media photograph using the provided image as the base.HUMAN SUBJECT — ABSOLUTE LOCK:- Keep the exact original person from the reference photo.- Do NOT regenerate, redraw, beautify, reshape, retouch, or replace the person.- Preserve the exact face, facial structure, facial proportions, skin texture, hairstyle, cap, clothing, body shape, pose, posture, hand position, and expression.- Preserve the original smile and the exact candid “looking back” expression.- The human must remain photorealistic and visually identical to the source photograph.- No AI beauty filter, no face alteration, no skin smoothing, no facial reconstruction.- The person is the photographic anchor of the composition.ENVIRONMENT — TRANSFORM INTO EDITORIAL DOODLE COLLAGE:Transform ONLY the surrounding environment into a sophisticated hand-drawn editorial collage.Keep the original garden composition recognizable:- agave plants- tall columnar cacti- trees- sky- garden path- surrounding foliageAdd expressive hand-drawn doodles and graphic interventions around the environment:- rough black marker outlines- cream/white chalk strokes- mustard-yellow hand-drawn lines- arrows- stars- crowns- circles- underlines- sketch marks- imperfect handwritten annotations- playful graphic symbols- subtle torn-paper textures- distressed ink textures- analog scrapbook elements- subtle grain and print imperfectionsThe doodles should interact with the plants and background naturally, almost as if an editorial photographer photographed a real location and then an artist drew directly over the photograph.IMPORTANT:The doodles must NOT cover, distort, redraw, or modify the human subject.Keep a clean visual separation around the person's face, body, clothing, and silhouette.TYPOGRAPHY:Add sparse, tasteful handwritten editorial notes integrated into the environment, such as:“GOOD DAY”“SAME PLACE / DIFFERENT MOOD”“JUST A BOY”“MORE LIFE HERE”“NATURE ALWAYS A GOOD IDEA”Use imperfect handwritten lettering rather than polished digital typography.ART DIRECTION:Youthful contemporary fashion-editorial aesthetic.Analog scrapbook + street editorial + botanical photography.Energetic but sophisticated.Controlled visual chaos.Premium Instagram campaign aesthetic.Natural photographic lighting.Strong depth and foreground layering.Organic composition rather than a symmetrical graphic layout.COLOR & FINISH:Keep the original natural photographic colors of the human and environment.Use black, warm cream, muted mustard yellow, and subtle off-white for the doodle elements.Add subtle film grain, paper texture, and analog print character.Avoid excessive saturation.FINAL RESULT:A highly photogenic editorial photograph that looks like a real candid outdoor fashion/lifestyle photograph enhanced with hand-drawn graphic artwork.MOST IMPORTANT RULE:THE HUMAN SUBJECT MUST REMAIN EXACTLY AS IN THE ORIGINAL PHOTO.ONLY THE ENVIRONMENT AND GRAPHIC ELEMENTS MAY BE STYLIZED",
    "width": "1024",
    "height": "1024",
    "samples": "1",
    "num_inference_steps": "30"
  })
})
.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-sft"                       # Find this model
modelslab models detail --id moss-image-sft                    # Get model details
modelslab generate image --prompt "..." --model moss-image-sft  # Generate with model
modelslab config set generation.default_model moss-image-sft   # 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-sft)
- [API Documentation](https://docs.modelslab.com)
- [ModelsLab Platform](https://modelslab.com)