capabilities.work

Generate And Describe Image

A 3-step recipe with Hyperbolic. We can help you get access.

Recipe · 3 steps

Generate And Describe Image

with Hyperbolic

Generate an image, caption it with a vision model, and narrate the caption.

  1. Generate a single image from the supplied prompt and capture its base64 bytes.
  2. Pass the generated image to a vision model as a data URI and ask for a one sentence caption.
  3. Convert the generated caption to speech audio and capture the base64 audio and its duration.
What you'll need to have ready
  • apiKey: Hyperbolic API key passed as a Bearer token.
  • imageModel: Image model name (e.g. SDXL1.0-base).
  • imagePrompt: The text-to-image prompt to render.
  • visionModel: Vision-capable chat model id (e.g. meta-llama/Llama-3.2-90B-Vision-Instruct).
  • language: TTS language code for narration.
  • speed: Speech speed between 0.5 and 2.0.
For your developer

Arazzo workflow hyperbolic-ai:generate-and-describe-image, 3 operations: generateImage → createChatCompletion → generateAudio. Specs on APIs.io.

More recipes with Hyperbolic

Chat To ImageChat To SpeechDiscover Model And ChatDiscover Model And CompleteImage Prompt QA LoopMultimodal StoryResearch Summarize And NarrateTool Calling Roundtrip