A WireMock simulation of Hugging Face's REST API — covering dataset validity and splits, metadata and sizes, Parquet listings, first rows, row paging, full-text search, filtering, and column statistics.
A shared instance of this template runs on a public endpoint — call it right now from curl, an SDK, or an AI agent. Responses are fixed examples and behavior can't be changed; rate limits apply.
Jump to the endpoints ↓Run this template as a mock API in WireMock Cloud for a private instance you control — edit stubs, add stateful scenarios, use dynamic response templating, and inspect requests in the log.
Open in WireMock Cloud (free) →Live sandbox · no signup, no API key
A running WireMock Cloud instance of this template, callable right now from a terminal, a script, or an AI agent.
Endpoints
GET First rows of a split [listFirstRows] - train
curl -s -X GET 'https://ygy7qk.wiremockapi.cloud/first-rows?dataset=stanfordnlp%2Fimdb&config=plain_text&split=train'GET First rows of a split [listFirstRows] - test
curl -s -X GET 'https://ygy7qk.wiremockapi.cloud/first-rows?dataset=stanfordnlp%2Fimdb&config=plain_text&split=test'GET First rows of a split [listFirstRows] - unsupervised
curl -s -X GET 'https://ygy7qk.wiremockapi.cloud/first-rows?dataset=stanfordnlp%2Fimdb&config=plain_text&split=unsupervised'Spec catalog — Dataset viewer server API (1.0) · 12 endpoints
| Method | Path | Summary |
|---|---|---|
| GET | /filter | Filter rows of a split |
| GET | /first-rows | First rows of a split |
| GET | /info | Get the metadata of a dataset. |
| GET | /is-valid | Check if a dataset is valid |
| GET | /opt-in-out-urls | Get the number of opted-in and opted-out image URLs in a dataset. |
| GET | /parquet | List of parquet files |
| GET | /presidio-entities | Get the number of rows containing Presidio entities in a dataset. |
| GET | /rows | A slice of rows of a split |
| GET | /search | Full-text search in the text columns of a split |
| GET | /size | Get the size of a dataset. |
| GET | /splits | List of splits |
| GET | /statistics | Descriptive statistics of a split's columns |
Unauthenticated WireMock Cloud demo sandbox for Hugging Face — not an official Hugging Face sandbox, and it returns simulated example data only.
The template replays real responses from datasets-server.huggingface.co, recorded for the public stanfordnlp/imdb dataset. Row requests page through real reviews with their true row_idx and the split's real total, 25,000 rows for train. Filtering the test split on "label"=1 starts at row 12,500, because imdb's splits are sorted by label. Code that previews datasets, builds sample pages or checks split sizes gets genuine data to work with.
Only stanfordnlp/imdb, in its plain_text config with the train, test and unsupervised splits. Any other dataset answers the real 401 the live service gives for something it can't see, which is also how it treats private and gated datasets.
They run over the 100 recorded rows of each split, so num_rows_total is the number of matches in that pool. Filters support one equality condition; other where expressions match nothing.
Paging keeps working. Rows beyond the recorded 100 repeat those rows in order while row_idx keeps counting, so loops that walk a whole split reach the end in the expected number of pages.
No, as with the live service for public datasets. Don't add a real token to requests against the shared sandbox.
Vendor names identify APIs represented by WireMock template sources. This page does not imply vendor endorsement, certification, partnership, or official integration status.