Skip to main content
Connect AI and machine learning services so your agent can generate text, run models, and synthesize audio as part of any workflow.
OpenAI lets your agent generate text, images, and embeddings using OpenAI’s models.What you can do:
  • Generate text completions and chat responses using GPT models
  • Create images with DALL·E
  • Produce vector embeddings for semantic search or similarity tasks
  • Transcribe audio with Whisper
Credentials: API key — find it in the OpenAI platform under API keys.Example prompt:
“Use OpenAI to summarize this 10-page PDF into a 5-bullet executive summary.”
Replicate lets your agent run machine learning models via the Replicate API.What you can do:
  • Run any publicly available model on Replicate
  • Pass custom inputs and retrieve model outputs
  • Run image, audio, video, and text models
Credentials: API token — find it in your Replicate account settings.Example prompt:
“Use Replicate to upscale this image to 4x resolution using the Real-ESRGAN model.”
ElevenLabs lets your agent generate realistic speech and clone voices.What you can do:
  • Convert text to speech using any available voice
  • Clone a voice from audio samples
  • Generate audio in multiple languages
Credentials: API key — find it in your ElevenLabs profile under Profile → API key.Example prompt:
“Use ElevenLabs to generate an audio version of this week’s newsletter using the Rachel voice.”