Sogni AI

AI Image automation

v1.2.6

Last updated Nov 3, 2025

n8n community node for Sogni AI image and video generation

281 Weekly Downloads

346 Monthly Downloads

View on NPM GitHub Repository

Included Nodes

Sogni AI

Description

n8n-nodes-sogni

Enhanced n8n Community Node for Sogni AI Image & Video Generation

Generate AI images and videos using Sogni AI Supernet directly in your n8n workflows with full ControlNet support for guided image generation and video generation capabilities.

This node pulls from your personal Sogni account—sign up for free to get 50 free Render credits per day. Under the hood, the project utilizes the @sogni-ai/sogni-client-wrapper, which is built on top of the official @sogni-ai/sogni-client SDK.

Example n8n workflow using the Sogni node

—

🆕 What’s New

🎬 Video Generation Support (v1.2.0)

Generate AI videos with customizable frames, FPS, and resolution
MP4 video output format
Automatic video download as binary data
Configurable video parameters (frames, guidance, steps)
Dedicated video model selection

📥 Automatic Image Download

Download generated images as binary data
Prevents 24-hour URL expiry issues
Proper MIME type handling
Enabled by default for reliability

🔑 Enhanced AppId Management

Auto-generates unique appId per execution
Prevents WebSocket socket collisions
Supports concurrent workflow runs
Manual override still available

⚙️ Improved Defaults

Default network: fast (quicker generation)
Timeout defaults by network when not set: fast = 60s, relaxed = 600s
Default token type: spark
Download images: enabled by default

✨ Full ControlNet Support

15 ControlNet types supported (canny, scribble, lineart, openpose, depth, and more)
Guide image generation with control images
Full parameter control (strength, mode, guidance timing)
See ControlNet Guide for details

—

Features

Resources & Operations

#### Image Resource

Generate: Create AI images with optional ControlNet guidance

#### Video Resource

Generate: Create AI videos with customizable parameters

#### Model Resource

Get All: List all available models
Get: Get specific model details

#### Account Resource

Get Balance: Check SOGNI and Spark token balance

—

Installation

Option 1: Community Nodes UI (Recommended)

1. In n8n, open Settings ▸ Community Nodes
2. Select Install
3. Enter n8n-nodes-sogni
4. Confirm the installation (restart n8n if prompted)

Option 2: Manual Installation

Run in your n8n installation directory
npm install n8n-nodes-sogni
Restart your n8n instance after installation

—

Configuration

1. Add Credentials

1. In n8n, go to Credentials
2. Click Add Credential
3. Search for “Sogni AI”
4. Enter your credentials:
– Username: Your Sogni account username
– Password: Your Sogni account password
– App ID: (Optional) Leave empty for auto-generation

2. Add Node to Workflow

1. Create or open a workflow
2. Click + to add a node
3. Search for “Sogni AI”
4. Select the node and configure

—

Basic Usage

> 💡 Tip: You can import example workflows directly into n8n! Create a new workflow, click the ⋮ (three dots) in the top right corner, select Import from File…, and choose a sample workflow from the ./examples folder.

Simple Image Generation

{
  "resource": "image",
  "operation": "generate",
  "modelId": "flux1-schnell-fp8",
  "positivePrompt": "A beautiful sunset over mountains",
  "network": "fast",
  "additionalFields": {
    "negativePrompt": "blurry, low quality",
    "steps": 20,
    "guidance": 7.5,
    "tokenType": "spark",
    "downloadImages": true
  }
}

ControlNet-Guided Generation

{
  "resource": "image",
  "operation": "generate",
  "modelId": "flux1-schnell-fp8",
  "positivePrompt": "A fantasy castle, magical, glowing",
  "network": "fast",
  "additionalFields": {
    "enableControlNet": true,
    "controlNetType": "canny",
    "controlNetImageProperty": "data",
    "controlNetStrength": 0.7,
    "controlNetMode": "balanced",
    "steps": 20,
    "downloadImages": true
  }
}

Video Generation

{
  "resource": "video",
  "operation": "generate",
  "videoModelId": "wanv2.2-14b-fp8t2v_lightx2v",
  "videoPositivePrompt": "A serene waterfall flowing through a lush green forest",
  "videoNetwork": "fast",
  "videoAdditionalFields": {
    "videoSettings": {
      "frames": 81,
      "fps": 16,
      "steps": 4,
      "guidance": 7.5
    },
    "output": {
      "downloadVideos": true,
      "outputFormat": "mp4",
      "width": 640,
      "height": 640
    },
    "advanced": {
      "tokenType": "spark",
      "timeout": 300000
    }
  }
}

—

ControlNet Types

All 15 ControlNet types are supported:

See ControlNet Guide for detailed usage instructions.

—

Parameters

Required Parameters

Optional Parameters (Additional Fields)

ControlNet Parameters

Video Generation Parameters

#### Optional Video Parameters

—

Example Workflows

See the examples directory for complete workflow JSON files:

1. Basic Image Generation – Simple text-to-image
2. Batch Processing – Generate multiple images
3. Dynamic Model Selection – Auto-select best model
4. Scheduled Generation – Daily automated images
5. Video Generation – AI video creation with customizable parameters
6. ControlNet Edge Detection – Structure-guided generation (coming soon)
7. ControlNet Pose Transfer – Pose-guided generation (coming soon)

—

Output

Image Generation Output

#### JSON Output

{
  "projectId": "ABC123...",
  "modelId": "flux1-schnell-fp8",
  "prompt": "A beautiful sunset...",
  "imageUrls": [
    "https://complete-images-production.s3-accelerate.amazonaws.com/..."
  ],
  "completed": true,
  "jobs": [
    {
      "id": "JOB123...",
      "status": "completed"
    }
  ]
}

#### Binary Output (when downloadImages = true)

image: First generated image
image_1: Second image (if multiple)
image_2: Third image (if multiple)
etc.

Video Generation Output

#### JSON Output

{
  "projectId": "VID123...",
  "modelId": "video-model-id",
  "prompt": "A cat playing...",
  "videoUrls": [
    "https://complete-videos-production.s3-accelerate.amazonaws.com/..."
  ],
  "completed": true,
  "jobs": [
    {
      "id": "JOB456...",
      "status": "completed"
    }
  ]
}

#### Binary Output (when downloadVideos = true)

video: First generated video
video_1: Second video (if multiple)
video_2: Third video (if multiple)
etc.

Binary data includes:

Proper MIME type (video/mp4)
Filename: sognivideo[projectId]_[index].[ext]
Full resolution video data

—

Tips & Best Practices

1. Network Selection

Fast Network:

– Uses SOGNI tokens
– Faster generation (seconds to minutes)
– Higher cost
– Best for: Time-sensitive applications

Relaxed Network:

– Uses Spark tokens
– Slower generation (minutes to hours)
– Lower cost
– Best for: Batch processing, scheduled jobs

2. Model Selection

Popular models:

flux1-schnell-fp8: Fast, high quality, 4 steps recommended
coreml-sogniartistv1_768: Artistic style
chroma-v.46-flash_fp8: Fast generation

Use “Get All Models” operation to see all available models.

3. Steps Configuration

Flux models: 4-8 steps (optimized for speed)
SD models: 15-30 steps (better quality)
ControlNet: 20-30 steps (more control)

4. ControlNet Usage

Start with strength 0.5 and adjust
Use balanced mode for most cases
Match ControlNet type to your control image
See ControlNet Guide for details

5. Image Download

Enable downloadImages to prevent URL expiry
URLs expire after 24 hours
Binary data is permanent in n8n
Recommended for production workflows

6. Timeout Configuration

Image – Fast network: 60,000ms (1 minute) usually enough
Image – Relaxed network: 600,000ms (10 minutes) recommended
Video – Fast network: 120,000ms (2 minutes) minimum
Video – Relaxed network: 1,200,000ms (20 minutes) recommended
Adjust based on complexity and model

7. Video Generation Tips

Frame Count: Start with 30 frames for quick tests, increase for longer videos
FPS: Use 30 fps for smooth motion, 10-15 fps for stylized/animated look
Resolution: Start with 512×512 for faster generation, increase as needed
Format: Currently only MP4 format is supported
Models: Look for models with “video”, “animation”, or “motion” in their names

—

Troubleshooting

“Insufficient funds” Error

Solution: Add more Spark or SOGNI tokens to your account

“Model not found” Error

Solution: Use “Get All Models” to see available models

“No binary data found” (ControlNet)

Solution:
1. Ensure previous node outputs binary data
2. Check the binary property name
3. Use “View” in n8n to inspect data

Workflow Times Out

Solution:

Use relaxed network for slower but more reliable generation
Increase timeout in Additional Fields
Split large batches into smaller chunks

Images Not Downloaded

Solution:

Check downloadImages is enabled
Verify network connectivity
Check n8n logs for download errors

—

Advanced Usage

Combining with Other Nodes

#### Discord Integration

Sogni Generate → HTTP Request (Discord Webhook)

#### Google Drive Storage

Sogni Generate → Google Drive (Upload File)

#### Social Media Posting

Sogni Generate → Twitter/Instagram API

#### Image Processing Pipeline

Load Image → Sogni ControlNet → Post-Processing → Save

Dynamic Prompts

Use expressions to generate dynamic prompts:

{{ "A " + $json.style + " image of " + $json.subject }}

Conditional ControlNet

Enable ControlNet based on conditions:

{{ $json.hasControlImage ? true : false }}

—

API Reference

Wrapper Library

This node uses the @sogni-ai/sogni-client-wrapper library. For standalone Node.js usage:

import { SogniClientWrapper } from '@sogni-ai/sogni-client-wrapper';
const client = new SogniClientWrapper({
  username: 'your-username',
  password: 'your-password',
  autoConnect: true,
});const result = await client.createProject({
  modelId: 'flux1-schnell-fp8',
  positivePrompt: 'A beautiful sunset',
  network: 'fast',
  tokenType: 'spark',
  waitForCompletion: true,
});

See @sogni-ai/sogni-client-wrapper for full API documentation.

—

Version History

v1.2.0 (Current)

🎬 Added full video generation support
📦 Updated @sogni-ai/sogni-client-wrapper to v1.2.0
🎥 MP4 video format support
⚙️ Configurable video parameters (frames, FPS, resolution)
📥 Automatic video download as binary data
🔍 Dedicated video model selection and filtering

v1.1.9

📝 Updated Sogni signup copy and highlighted ControlNet positioning

v1.1.8

🆕 Refreshed installation instructions and Sogni account links
📚 Added references to Sogni platform, docs, and SDK packages

v1.1.6

⚡ Changed default network from “relaxed” to “fast” for quicker generation
📝 Documentation updates

v1.1.5

🔧 Minor bug fixes and improvements
📝 Documentation updates

v1.1.0-1.1.4

✨ Added full ControlNet support (15 types)
📥 Added automatic image download
🔑 Enhanced appId auto-generation
⚙️ Improved default values
📚 Added ControlNet guide

v1.0.0

Initial release
Basic image generation
Model and account operations
Size presets support

—

Resources

ControlNet Guide – Complete ControlNet usage guide
Example Workflows – Ready-to-use workflow examples
Sogni AI – Platform overview and product updates
Sogni SDK Docs – Official SDK documentation
Sogni Docs – Platform guides and API references
Integration Guide – Complete integration guide

—

Support

For issues or questions:
1. Check this README
2. Review the ControlNet Guide
3. Check example workflows
4. Submit an issue on GitHub

—

License

MIT License – See LICENSE file for details

—

Credits

Built with:

@sogni-ai/sogni-client – Official Sogni SDK
@sogni-ai/sogni-client-wrapper – Enhanced wrapper library
n8n – Workflow automation platform

—

Ready to generate amazing AI images with ControlNet in your n8n workflows! 🎨✨