Hugging Face Dataset Info (Features, Splits, Sizes)
huggingfacedatasetsserver/get-info
Returns the Dataset Viewer's info for one public Hugging Face dataset: for each configuration (or for the one `config` asked for) the feature schema (column names and types, class label names), the splits with their example counts and byte sizes, the download and dataset sizes, and the description, citation, homepage and licence text if the authors filled them in. Strings are shortened to 500 characters, lists to 20 entries and objects to 100 keys, and the answer carries `clipped: true` when anything was cut. Each dataset has its own licence, set by its authors on the dataset card and not returned by this tool; huggingfacehub/get-dataset returns it. Hugging Face Dataset Viewer data, served by Eden Engine; not an official Hugging Face product.
Price
$0.0026 per call
You pay the price shown. If a call turns out to cost more than its quote, you pay the actual price, capped at twice the quote. No monthly minimum. Top-ups start at $0.50.
Data source
Data from the Hugging Face Dataset Viewer API (https://huggingface.co/docs/dataset-viewer). Each dataset carries its own licence, set by its authors on the dataset card. Served by Eden Engine; not an official Hugging Face product.
Use it
Five steps: search, inspect, learn, run, settle. No per-tool setup; your agent reaches this tool through the meta-tools it already has.
Connect once and your AI can find and run this tool, and any other it needs.
// 1. search: find the tool
omnial_search({ query: "what you need" })
// 2. inspect: price and input schema
omnial_inspect({ tool: "huggingfacedatasetsserver/get-info" })
// 3. learn: read the usage guide at runtime
omnial_learn({ tool: "huggingfacedatasetsserver/get-info" })
// 4. run: price the exact call first, then run it
omnial_execute({ tool: "huggingfacedatasetsserver/get-info", input: { /* see input_schema */ }, dry_run: true })
omnial_execute({ tool: "huggingfacedatasetsserver/get-info", input: { /* ... */ } })
// 5. settle: the call is charged when it finishes, nothing to doNot connected yet? Point your MCP client at https://mcp.eden-engine.com/mcp, then copy a config for your client.
Pricing
Every charge this tool can incur.
| Charge | Price |
|---|---|
Per call | $0.0026 |
- Updated
- Oct 25, 2026
Input and output schemas
Input schema
{
"type": "object",
"required": [
"dataset"
],
"properties": {
"config": {
"type": "string",
"pattern": "^[A-Za-z0-9][A-Za-z0-9._+ -]{0,99}$",
"description": "One config name from get-splits. Omit it for every configuration of the dataset."
},
"dataset": {
"type": "string",
"pattern": "^[A-Za-z0-9][A-Za-z0-9._-]*(/[A-Za-z0-9][A-Za-z0-9._-]*)?$",
"maxLength": 100,
"minLength": 3,
"description": "Public dataset repo id, owner/name, for example stanfordnlp/imdb. Older canonical datasets without an owner (for example squad) are accepted."
}
},
"additionalProperties": false
}Output schema
{
"type": "object",
"required": [
"dataset_info"
],
"properties": {
"failed": {
"type": "array",
"items": {
"type": "object",
"additionalProperties": true
},
"description": "Configs whose data is still being computed."
},
"clipped": {
"type": "boolean",
"description": "True when the hook shortened a string, list or object; strings end with a note giving the count dropped."
},
"partial": {
"type": "boolean"
},
"pending": {
"type": "array",
"items": {
"type": "object",
"additionalProperties": true
},
"description": "Configs whose data is still being computed."
},
"dataset_info": {
"type": "object",
"additionalProperties": true
}
},
"description": "The Dataset Viewer's own answer, bounded by the hook (see `clipped`). With `config` the `dataset_info` object is that config's info; without it, a map keyed by config name.",
"additionalProperties": true
}