Search model metadata in the Hub. Inference uses AI > Models.
Explore 2 workflow actions for Hugging Face, including account permissions and operation limits.
2 read actions0 write actions
WORKFLOW ACTIONS
What you can configure.
Choose an action in the workflow builder, connect an account, and map its inputs. These actions are available in the catalog; your account access and a successful test determine whether they work for your setup.
Get model
Read
Read metadata for one model repository without fetching its files.
Permissions and limits
Account permissions
Hugging Face token allowed to read this model
Operation limits
Model files, gated access requests, model card text and inference output are excluded.
Search models
Read
Read one bounded set of model identities and popularity metrics.
Permissions and limits
Account permissions
Hugging Face token allowed to read matching models
Operation limits
A single bounded result set. Public models can appear regardless of the key's private repository permissions. Listing a model does not grant inference access.
SETUP
Start with a controlled test.
01Connect the intended account
Use the account and permissions required for your process. Review the connection check before configuring a workflow action.
02Review inputs and access
Select a listed operation and map its required inputs. Use test records and add a human approval before writes that need review.
03Confirm the result
Inspect step outputs and confirm the result in Hugging Face. A saved workflow is not a completed provider action.