Linking Hugging Face as a source
This source is currently in alpha. The interface and available tables may change.
The Hugging Face connector syncs the repositories owned by a Hub namespace – models, datasets, and Spaces – into PostHog, so you can analyze your Hub activity alongside your product data.
Prerequisites
You need a Hugging Face account and a User Access Token. A read token is enough for public repositories. To include private repositories, the token needs read access to the namespace's repository contents, plus read-org if you are syncing an organization.
Adding a data source
- In PostHog, go to the Sources tab of the data pipeline section.
- Click + New source and click Link next to this source.
- Enter your credentials (see Configuration below) and click Next.
- Select the tables you want to sync, choose a sync method and frequency, then click Import.
Once the syncs are complete, you can start querying this data in PostHog.
You need two things to connect Hugging Face:
- Access token – create one in your Hugging Face access token settings. It starts with
hf_. - Username or organization – the namespace whose models, datasets, and Spaces you want to sync (for example your username, or an organization you belong to).
Sync modes
Each table can be synced in one of several modes, depending on what the source supports:
- Webhook (when available) – the source pushes changes to PostHog in real time. Fastest freshness, lowest ongoing cost, and the only mode that reliably captures updates and deletes.
- Incremental – only new or updated rows are synced on each run, using a cursor field (such as an
updated_attimestamp). Cheaper than a full refresh, but deletes aren't captured. - Append only – new rows are appended using a cursor field; existing rows are never updated. Ideal for immutable, append-only tables like event logs.
- Full refresh – the whole table is reloaded on every sync. Use it when a table has no reliable cursor or when you need deletions reflected.
See sync methods for a full explanation of how each mode works and how to choose between them.
The Hugging Face Hub has no server-side timestamp filter on its list endpoints, so every table is synced as a full refresh. Each sync walks the namespace's repositories newest-last and merges them on their repository id.
Configuration
| Option | Type | Required |
|---|---|---|
Access token | password | Yes |
Username or organization | text | Yes |
Supported tables
| Table | Description | Sync method | Incremental field | Primary key |
|---|---|---|---|---|
models | A model repository on the Hugging Face Hub, owned by the connected namespace. | Full refresh | — | — |
datasets | A dataset repository on the Hugging Face Hub, owned by the connected namespace. | Full refresh | — | — |
spaces | A Space (hosted app) repository on the Hugging Face Hub, owned by the connected namespace. | Full refresh | — | — |
Troubleshooting
If the connection fails with an authentication error, confirm the access token is still valid in your Hugging Face settings and that it grants read access to the namespace you entered. Regenerate the token and reconnect if it has been revoked.
If your sync is failing or data looks wrong, see the Data warehouse troubleshooting guide. If that doesn't help, contact support – we're happy to help.