Skip to main content

Documentation Index

Fetch the complete documentation index at: https://next.gentrace.ai/docs/llms.txt

Use this file to discover all available pages before exploring further.

The datasets SDK provides programmatic access to Gentrace datasets. Datasets organize test cases and are associated with pipelines for evaluation purposes.

​
Basic usage

import { init, datasets } from 'gentrace';

init({
  apiKey: process.env.GENTRACE_API_KEY,
});

// List datasets
const datasetList = await datasets.list();

// Access the datasets
for (const dataset of datasetList.data) {
  console.log(dataset.name);
  console.log(dataset.description);
}

​
Overview

The SDK is built by Stainless and provides type-safe access to Gentrace entities. The datasets object exposes methods to create, retrieve, update, and list datasets.

​
Dataset structure

Each dataset contains:
​
id
string
required
Unique identifier (UUID)
​
name
string
required
Human-readable name for the dataset
​
description
string
Optional description of the dataset’s purpose
​
pipelineId
string
required
UUID of the associated pipeline
​
archivedAt
string
Timestamp when archived (null if active)
​
createdAt
string
required
Creation timestamp
​
updatedAt
string
required
Last update timestamp

​
Resource methods

​
Create a dataset

const dataset = await datasets.create({
  name: 'Production feedback - Q1 2025',
  description: 'Negative user feedback from production',
  pipelineId: 'pipeline-uuid', // or use pipelineSlug
});

​
Retrieve a dataset

const dataset = await datasets.retrieve('dataset-id');
console.log(dataset.name);

​
Update a dataset

const dataset = await datasets.update('dataset-id', {
  name: 'Updated dataset name',
  isArchived: false,
  isGolden: true, // Mark as golden dataset
});

​
List with filters

// Filter by pipeline
const datasetList = await datasets.list({
  pipelineId: 'pipeline-id',
  archived: false, // Exclude archived datasets
});

​
Relationship with test cases

Datasets contain test cases. To work with test cases within a dataset, use the test cases SDK: 
// Get test cases for a dataset
const testCasesList = await testCases.list({
  datasetId: dataset.id,
});

​
See also

  • Test Cases - Working with test cases within datasets
  • Pipelines - Managing pipelines that contain datasets
  • evalDataset() - Running evaluations against datasets