Skip to main content

Replicate AI

Run AI models on Replicate's cloud platform for image generation, text processing, and more.

Overview​

The Replicate AI package provides integration with Replicate's AI model platform. Use it when you need to run machine learning models for image generation, text processing, audio generation, or other AI tasks without managing infrastructure.

Key Features​

  • Model Execution - Run any model from Replicate's library
  • Predictions - Create, get, list, and cancel predictions
  • Training - Train custom models
  • Async Processing - Handle long-running predictions

Available Nodes​

  • Connect - Authenticate with Replicate API
  • Run Model - Run a model and wait for results
  • Create Prediction - Start an async prediction
  • Get Prediction - Check prediction status and get results
  • List Prediction - List all predictions
  • Cancel Prediction - Cancel a running prediction
  • Get Model - Get information about a model
  • Create Training - Start model training
  • Get Training - Check training status

When to Use This Package​

  • Image Generation: Generate images with Stable Diffusion, DALL-E, etc.
  • Text to Speech: Convert text to audio
  • Image Processing: Upscale, edit, or transform images
  • Language Models: Run LLMs for text generation
  • Custom AI Tasks: Use specialized models from Replicate

Typical Workflow​

Synchronous​

  1. Connect with API token
  2. Run Model with inputs
  3. Get results directly

Asynchronous​

  1. Connect with API token
  2. Create Prediction to start long-running task
  3. Get Prediction to poll for completion
  4. Process results when ready

Use Cases​

  • Generate marketing images from text descriptions
  • Upscale low-resolution images
  • Transcribe audio files
  • Generate voiceovers from text
  • Run custom trained models