​
What can you do with it?

The /replicate command enables you to run various AI models through the Replicate platform. You can generate images or speech from text, and so much more!

​
How to use it?

​
Basic Command Structure

/replicate [task description]

​
Parameters

Required:

  • authorId - The author ID of the model (e.g., black-forest-labs, minimax, 851-labs)
  • modelId - The model ID (e.g., flux-schnell, speech-02-turbo, background-remover)
  • input - Input parameters specific to the model

Optional:

  • version - Specific model version to use
  • fileLinksExpireInDays - How long generated files remain accessible (1-7 days, defaults to 7)
  • fileLinksExpireInMinutes - How long generated files remain accessible in minutes (takes precedence over days)

​
Response Format

The command returns:

{
  "output": [
    {
      "url": "https://generated-file-url",
      "mimeType": "image/webp or audio/wav"
    }
  ]
}

Note: All generated files are automatically saved to your “Multimedia Artifact” file store and will be accessible for the duration specified by the fileLinksExpireInDays parameter.

​
Examples

​
Basic Image Generation

/replicate generate an image 
authorId: black-forest-labs
modelId: flux-schnell
prompt: A modern office space with plants
aspect_ratio: 16:9

Generates a quick image of an office space using the fast Flux model.

​
Text-to-Speech

/replicate text to speech
authorId: minimax
modelId: speech-02-turbo
text: Welcome to our automated customer service. How can I help you today?
voice_id: Friendly_Person
speed: 1.0
emotion: happy
fileLinksExpireInDays: 7

Generates happy speech audio with a friendly voice that remains accessible for 7 days.

​
Background Removal

/replicate remove background
authorId: 851-labs
modelId: background-remover
image: https://example.com/product-photo.jpg  (or select from your files)
format: png

Removes the background from a product photo and returns a transparent PNG.

​
Text-Based Image Editing

/replicate edit image
authorId: black-forest-labs
modelId: flux-kontext-max
prompt: Add a sunset sky background to this image
input_image: https://example.com/portrait.jpg (or select from your files)
aspect_ratio: match_input_image
output_format: png

Uses the premium flux-kontext-max model to edit an existing image with text instructions, providing high-quality results with improved typography generation.

​
Notes

Supported Content Types:

  • Images: Generate, edit, and transform images
  • Audio: Generate speech and audio processing
  • Video: NOT SUPPORTED

​
Model Parameters

​
Text-to-Speech (minimax/speech-02-turbo)

  • text (required): Text to convert to speech (max 5000 chars). Use <#x#> for pause control (0.01-99.99s)
  • pitch: Speech pitch (-12 to 12, default: 0)
  • speed: Speech speed (0.5 to 2, default: 1)
  • volume: Speech volume (0 to 10, default: 1)
  • bitrate: Bitrate (32000, 64000, 128000, 256000, default: 128000)
  • channel: Audio channels (“mono”, “stereo”, default: “mono”)
  • emotion: Speech emotion (“auto”, “neutral”, “happy”, “sad”, “angry”, “fearful”, “disgusted”, “surprised”, default: “auto”)
  • voice_id: Voice ID (default: “Wise_Woman”). Options: Wise_Woman, Friendly_Person, Inspirational_girl, Deep_Voice_Man, Calm_Woman, Casual_Guy, Lively_Girl, Patient_Man, Young_Knight, Determined_Man, Lovely_Girl, Decent_Boy, Imposing_Manner, Elegant_Man, Abbess, Sweet_Girl_2, Exuberant_Girl
  • sample_rate: Sample rate (8000, 16000, 22050, 24000, 32000, 44100, default: 32000)
  • language_boost: Language enhancement (“None”, “Automatic”, “Chinese”, “Chinese,Yue”, “English”, “Arabic”, “Russian”, “Spanish”, “French”, “Portuguese”, “German”, “Turkish”, “Dutch”, “Ukrainian”, “Vietnamese”, “Indonesian”, “Japanese”, “Italian”, “Korean”, “Thai”, “Polish”, “Romanian”, “Greek”, “Czech”, “Finnish”, “Hindi”, default: “None”)
  • english_normalization: Enable English text normalization for better number reading (boolean, default: false, slightly increases latency)

​
Image Generation (black-forest-labs/flux-schnell)

  • prompt (required): Prompt for generated image
  • aspect_ratio: Aspect ratio (“1:1”, “16:9”, “21:9”, “3:2”, “2:3”, “4:5”, “5:4”, “3:4”, “4:3”, “9:16”, “9:21”, default: “1:1”)
  • num_outputs: Number of outputs to generate (1 to 4, default: 1)
  • num_inference_steps: Number of denoising steps (1 to 4, default: 4). Lower steps = faster but lower quality
  • seed: Random seed for reproducible generation (integer, optional)
  • output_format: Format of output images (“webp”, “jpg”, “png”, default: “webp”)
  • output_quality: Quality when saving images (0 to 100, default: 80). Not relevant for PNG outputs
  • disable_safety_checker: Disable safety checker for generated images (boolean, default: false)
  • go_fast: Run faster predictions with fp8 quantized model (boolean, default: true). Note: outputs not deterministic when enabled
  • megapixels: Approximate megapixels for generated image (“1”, “0.25”, default: “1”)

​
Background Removal (851-labs/background-remover)

  • image (required): Input image (URI format)
  • threshold: Threshold for hard segmentation (0.0-1.0, default: 0). If 0.0, uses soft alpha
  • reverse: If true, remove the foreground instead of the background (boolean, default: false)
  • background_type: Background type (default: “rgba”). Options: ‘rgba’, ‘map’, ‘green’, ‘white’, [R,G,B] array, ‘blur’, ‘overlay’, or path to an image
  • format: Output format (default: “png”). Examples: png, jpg

​
Text-Based Image Editing (black-forest-labs/flux-kontext-max)

  • prompt (required): Text description of what you want to generate, or instruction on how to edit the given image
  • input_image: Optional: Image to use as reference. Must be jpeg, png, gif, or webp (URI format, nullable)
  • aspect_ratio: Aspect ratio (“match_input_image”, “1:1”, “16:9”, “9:16”, “4:3”, “3:4”, “3:2”, “2:3”, “4:5”, “5:4”, “21:9”, “9:21”, “2:1”, “1:2”, default: “match_input_image”)
  • prompt_upsampling: Automatic prompt improvement (boolean, default: false)
  • seed: Random seed for reproducible generation (integer, nullable, optional)
  • output_format: Output format (“jpg”, “png”, default: “png”)
  • safety_tolerance: Safety tolerance (0-6, default: 2). 0 is most strict, 6 is most permissive. Max 2 when input images are used