---
title: "Get a dataset"
method: GET
path: "/datasets/{dataset_id}"
tags: ["datasets"]
---

# Get a dataset

`GET /datasets/{dataset_id}`

Returns a dataset by ID, including the total number of files and a list of each file's name and size in bytes.

## Path parameters

- `dataset_id` string, required

## Headers

- `Api-Key` string, required

## Response `200`

Dataset retrieved successfully

- GetDatasetResponse — Detailed view of a dataset, including its files.
  - `dataset` Dataset, required — A dataset for organizing training images.
    - `dataset_id` string, required — Unique identifier for the dataset.
    - `name` string, required — Display name of the dataset.
    - `user_id` string, required — ID of the user who owns this dataset.
    - `creation_time` string, date-time, required — When the dataset was created.
  - `file_count` integer, required — Total number of files in the dataset.
  - `files` DatasetFile[], required — List of files in the dataset.
    - `file_name` string, nullable, required — Filename of the file in the dataset.
    - `file_size_bytes` integer, nullable — Size of the file in bytes, if available.
    - `caption` string, nullable — Caption applied to the file, if any. A v4 structured caption is returned as a JSON-encoded string; a plain-text caption (from a .txt sidecar) is returned unchanged. Only present when the file has a caption.
  - `custom_model_ids` string[], required — IDs of all custom models trained from this dataset. Empty array if no model has been trained yet.

## Other responses

- `401` — Unauthorized
- `404` — Dataset not found

---

[API](https://skmtc.dev/ideogram/apis/api-reference.md) · [All operations](https://skmtc.dev/ideogram/apis/api-reference/llms.txt) · [OpenAPI document](https://skmtc-service-production.skmtc.workers.dev/v1/apis/ideogram/api-reference/revisions/d4cec35f719f/schema)