Skip to main content
Run AI models like FLUX, Stable Diffusion, LLaMA, Whisper and thousands more via a simple API

​
Capabilities

Once connected, your agent can run any of the thousands of open-source AI models hosted on Replicate.
  • Generate images with FLUX, Stable Diffusion, and more
  • Run language models (LLaMA, Mistral, etc.)
  • Transcribe audio with Whisper
  • Generate music and audio
  • Create and edit videos
  • Upscale and restore images
  • Remove image backgrounds
  • Fine-tune models on your own data
  • Search thousands of open-source models
  • Poll async predictions for long-running jobs

​
Setup Guide

Replicate uses a simple API token for authentication. No OAuth or complex setup required.

​
1. Create a Replicate account

Go to replicate.com and sign up with GitHub or email.

​
2. Get your API token

Go to Account → API Tokens and click “Create token”. Your token starts with r8_. Copy it immediately — it won’t be shown again.

​
3. Store the token in Environment

Go to Environment and create a new secret:
  • Name: REPLICATE_API_TOKEN
  • Value: Paste the token (starts with r8_)

​
4. Grant access to instances

Choose global access (all instances) or grant to specific instances only.
Billing note: Replicate charges per second of compute time. Check model pricing on the Replicate website before running expensive models. The agent will confirm with you before executing predictions.
Replicate hosts thousands of models. Here are some commonly used ones. Ask the agent to search for more.
CategoryModelDescription
Image Generationblack-forest-labs/flux-schnellFast high-quality images
Image Generationblack-forest-labs/flux-devHigher quality, slower
Image Generationstability-ai/sdxlStable Diffusion XL
Language Modelmeta/meta-llama-3-70b-instructText generation & chat
Speech to Textopenai/whisperAudio transcription
Musicmeta/musicgenMusic generation from text
Videominimax/video-01-liveVideo generation
Image Upscalenightmareai/real-esrganImage upscaling 4x
Background Removallucataco/remove-bgRemove image backgrounds

​
Usage Examples

Ask your agent to generate images, transcribe audio, and more. Image generation
“Generate an image of a sunset over mountains using FLUX”
“Create a portrait in the style of Van Gogh using Stable Diffusion”
Audio transcription
“Transcribe this audio file using Whisper”
“Generate a music track from the description: upbeat electronic dance music”
Image processing
“Upscale this image to 4x resolution”
“Remove the background from this product photo”
Discover models
“Search Replicate for text-to-speech models”
“Show me the input parameters for black-forest-labs/flux-dev”