Fine-tuning corpora
Domain text and structured records collected with provenance.
← All integrations
Import.io + Hugging Face
Web datasets on the Hugging Face Hub
Load the latest run with the datasets library and push it to a private Hub repo, versioned and ready for training or evaluation.
Python
import os
from datasets import load_dataset
ds = load_dataset("json", data_files=os.environ["IMPORTIO_JSON_URL"], split="train") # NDJSON
ds.push_to_hub("your-org/retail-prices-2026", private=True)How it connects.
Source
The web
Any public page, rendered, through managed proxies
Engine
Import.io
Rendering, proxy routing, captcha handling, extraction
Path
Data endpoint
Latest successful run, CSV or JSON
Destination
Hugging Face
Extracted records pushed as a versioned dataset to the Hub.
Also connects via managed delivery.
What teams do with it.
Evaluation sets that stay current
Rebuild eval sets from today’s pages so models are tested on the live web.
Team-wide dataset access
Private repos with versions, dataset cards and access control.
Set it up.
pip install datasets.- Log in with
huggingface-cli login. - Load the latest-run JSON and
push_to_hub.
Questions.
Does Import.io provide training data directly?
Yes. Managed Services build training and evaluation datasets to your specification, with schemas and provenance, and deliver them into your storage.
Do endpoint downloads count as queries?
No. The latest-run CSV, JSON and Google Sheets endpoints return data your extractor already collected, so they don’t count toward your plan. Only the Live Query API runs an extraction per request.
How fresh is the data?
As fresh as your schedule. Run the extractor hourly, daily or weekly and the endpoint always serves the latest successful run.
Who sets up managed delivery?
Your Managed Services team. They agree destination, format, schedule and schema with you, then build, run, monitor and QA the feed under an SLA.
Which destinations can Managed Services deliver into?
Amazon S3, Google Cloud Storage, Azure Blob Storage, SFTP, Snowflake, BigQuery, API, email and Slack alerts. Aperture delivers into the same destinations.
Related integrations.
EndpointMCP
Pinecone
Extracted pages embedded into a vector index for RAG.
Setup 15 minEndpointREST APIPython & pandas
Latest-run data into a DataFrame in three lines.
Setup 5 minS3 / SFTPManagedDatabricks
Auto Loader picks up every delivery into a Delta table.
Setup 15 minS3 / SFTPManagedAmazon S3
Every run delivered into your bucket, on schedule.
Setup 10 min