Run inference on 100,000+ open ML models for NLP, vision, and audio tasks
Hugging Face currently provides 8 actions a workflow can run. Use those building blocks to connect product and engineering activity to the customers and work it affects.
Document Question Answering
Answer questions from document images using Hugging Face models
Language Translation
Translate text between languages using specialized Hugging Face translation models
Text Classification
Classify text into categories using Hugging Face models - includes zero-shot classification for custom categories
Text Summarization
Generate abstractive summaries of long text using Hugging Face models - optimized for business content
Chat Completion
Generate assistant replies using chat-style LLMs - perfect for FAQ bots, support agents, and content generation
Create Image
Generate stunning images from text prompts using state-of-the-art diffusion models - perfect for marketing, product design, and creative content
Object Detection
Detect and locate objects in images with precise bounding boxes - perfect for inventory management, content moderation, and automated tagging
Image Classification
Classify images with pre-trained models or custom categories - perfect for content moderation, automated tagging, and smart asset management
These recipes are generated from the triggers, actions, and categories currently available for this piece. Use them as starting points, then choose the exact fields, conditions, and records your workflow should update.
Use a change in Coherence to run a supported Hugging Face action, so the next step follows the record instead of relying on someone to copy it.
Starts with
A Coherence record or workflow changes
Finishes with
Document Question Answering
Combine Hugging Face with another integration and Coherence records to connect product and engineering activity to the customers and work it affects.
Starts with
A Coherence workflow starts
Finishes with
Document Question Answering, then preserve the outcome in Coherence
The exact authentication fields depend on Hugging Face. The workflow structure stays straightforward:
Authorize the Hugging Face account and permissions the workflow needs.
Select one of the current triggers or actions listed above and map its fields.
Run a small test, inspect the result in both systems, and only then enable the workflow.
Explore apps that share a category with Hugging Face or provide a complementary trigger or action.
Hugging Face currently provides 8 actions a workflow can run. Use those building blocks to connect product and engineering activity to the customers and work it affects.
Hugging Face currently lists 0 triggers and 8 actions in the integration catalog. The supported list is refreshed from the pieces available in Coherence, so review the capability names on this page before building a workflow.
Yes. A workflow can combine supported Hugging Face triggers or actions with Coherence records and other available integrations. The exact flow depends on the authentication, permissions, triggers, and actions exposed by each connected app.
The listed triggers and actions are designed for visual workflows. If the pre-built integration does not expose a capability you need, HTTP and webhook steps can cover APIs that your team is authorized to use.
Start automating in minutes. No credit card required.
