Hugging Face

What you can do

8 actions, each one call from a workflow or a server function. Triggers come from your app: a row changing, a schedule or a webhook.

• Image Classificationimage_classification
  Classify images with pre-trained models or custom categories - perfect for content moderation, automated tagging, and smart asset management
• Text Classificationtext_classification
  Classify text into categories using Hugging Face models - includes zero-shot classification for custom categories
• Chat Completionchat_completion
  Generate assistant replies using chat-style LLMs - perfect for FAQ bots, support agents, and content generation
• Create Imagecreate_image
  Generate stunning images from text prompts using state-of-the-art diffusion models - perfect for marketing, product design, and creative content
• Document Question Answeringdocument_question_answering
  Answer questions from document images using Hugging Face models
• Language Translationlanguage_translation
  Translate text between languages using specialized Hugging Face translation models
• Object Detectionobject_detection
  Detect and locate objects in images with precise bounding boxes - perfect for inventory management, content moderation, and automated tagging
• Text Summarizationtext_summarization
  Generate abstractive summaries of long text using Hugging Face models - optimized for business content

Set it up

About two minutes, once per project or once for the whole workspace.

1. Open Integrations in the builder

   Backend panel, Integrations tab, then Hugging Face. To connect it for every project, use Workspace settings, Integrations.

2. Add your credentials

   An API key (or token) from your Hugging Face account, pasted once. It is stored encrypted and only read on the server when an action runs.

   • API key From Hugging Face's settings
3. Call an action

   From a workflow, by integration id and action name. The call answers with success, output and, when it fails, the reason.

   // .workflow.js (native ctx)
   export async function run(ctx) {
     const res = await ctx.integrations.invoke("hugging_face", "image_classification", {
       "classificationMode": "standard",
       "useCase": "moderation",
       "model": "<model>",
       "imageSource": "upload",
       "imageFile": "<image file>",
       "imageUrl": "<image url>",
       "customCategories": []
     });
     if (!res.success) throw new Error(res.error);
     ctx.log.info("Hugging Face answered", { output: res.output });
   }

More in AI and ML

• ABAmazon Bedrock
• AOAzure OpenAI
• Claude
• CoCohere
• Deepgram
• DeepSeek
• All of AI and ML