Skip to main content
Datasets enable you to perform repeatable evaluations over time using consistent data. Datasets are made up of examples, which store inputs, outputs, and optionally, reference outputs. This page outlines the various methods for creating and managing datasets in the UI.

​
Create a dataset and add examples

The following sections explain the different ways you can create a dataset in LangSmith and add examples to it. Depending on your workflow, you can manually curate examples, automatically capture them from tracing, import files, or even generate synthetic data:

​
Manually from a tracing project

A common pattern for constructing datasets is to convert notable traces from your application into dataset examples. This approach requires that you have configured tracing to LangSmith.
A technique to build datasets is to filter the most interesting traces, such as traces that were tagged with poor user feedback, and add them to a dataset. For tips on how to filter traces, refer to the Filter traces guide.
There are two ways to add data manually from a tracing project to datasets. Navigate to Tracing Projects and select a project.
  1. Multi-select runs from the runs table. On the Runs tab, multi-select runs. At the bottom of the page, click Add to Dataset.
  2. On the Runs tab, select a run from the table. On the individual run details page, select Add to -> Dataset in the top right corner. When you select a dataset from the run details page, a modal will pop up letting you know if any transformations were applied or if schema validation failed. You can then optionally edit the run before adding it to the dataset.

​
Automatically from a tracing project

You can use run rules to add traces automatically to a dataset based on certain conditions. For example, you could add all traces that are tagged with a specific use case or have a low feedback score.

​
From examples in an annotation queue

If you rely on subject matter experts to build meaningful datasets, use annotation queues to provide a streamlined view for reviewers. Human reviewers can optionally modify the inputs/outputs/reference outputs from a trace before it is added to the dataset.
You can optionally configure annotation queues with a default dataset, though you can add runs to any dataset by using the dataset switcher on the bottom of the screen. Once you select the right dataset, click Add to Dataset or hit the hot key D to add the run to it. Any modifications you make to the run in your annotation queue will carry over to the dataset, and all metadata associated with the run will also be copied.
You can also set up rules to add runs that meet specific criteria to an annotation queue using automation rules.

​
From the Playground

On the Playground page:
  1. Select Set up Evaluation.
  2. Click +New if you’re starting a new dataset or select from an existing dataset.
    Creating datasets inline in the Playground is not supported for datasets that have nested keys. In order to add/edit examples with nested keys, you must edit from the datasets page.
  3. Edit the examples:
    • Use +Row to add a new example to the dataset.
    • Delete an example using the ⋮ dropdown on the right-hand side of the table.
    • If you’re creating a reference-free dataset, remove the Reference Output column using the x button in the column. Note that this action is not reversible.

​
Import a dataset from a CSV or JSONL file

On the Datasets & Experiments page, click +New Dataset, then Import an existing dataset from CSV or JSONL file.

​
Create a new dataset from the datasets & experiments page

  1. Navigate to the Datasets & Experiments page from the left-hand menu.
  2. Click + New Dataset.
  3. On the New Dataset page, select the Create from scratch tab.
  4. Add a name and description for the dataset.
  5. (Optional) Create a dataset schema to validate your dataset.
  6. Click Create, which will create an empty dataset.
  7. To add examples inline, on the dataset’s page, go to the Examples tab. Click + Example.
  8. Define examples in JSON and click Submit. For more details on dataset splits, refer to Create and manage dataset splits.

​
Add synthetic examples created by an LLM

If you have existing examples and a schema defined on your dataset, when you click + Example there is an option to Add AI-Generated Examples. This will use an LLM to create synthetic examples. In Generate examples, do the following:
  1. Click API Key in the top right of the pane to set your OpenAI API key as a workspace secret. If your workspace already has an OpenAI API key set, you can skip this step.
  2. Select : Toggle Automatic or Manual reference examples. You can select these examples manually from your dataset or use the automatic selection option.
  3. Enter the number of synthetic examples you want to generate.
  4. Click Generate. The AI-Generated Examples configuration window. Selections for manual and automatic and number of examples to generate.
  5. The examples will appear on the Select generated examples page. Choose which examples to add to your dataset, with the option to edit them before finalizing. Click Save Examples.
  6. Each example will be validated against your specified dataset schema and tagged as synthetic in the source metadata. Select generated examples page with generated examples selected and Save examples button.

​
Manage a dataset

​
Create a dataset schema

LangSmith datasets store arbitrary JSON objects. We recommend (but do not require) that you define a schema for your dataset to ensure that they conform to a specific JSON schema. Dataset schemas are defined with standard JSON schema, with the addition of a few prebuilt types that make it easier to type common primitives like messages and tools. Certain fields in your schema have a + Transformations option. Transformations are preprocessing steps that, if enabled, update your examples when you add them to the dataset. For example, the convert to OpenAI messages transformation will convert message-like objects, like LangChain messages, to OpenAI message format. For the full list of available transformations, refer to the Dataset transformations reference.
If you plan to collect production traces in your dataset from LangChain ChatModels or from OpenAI calls using the LangSmith OpenAI wrapper, we offer a prebuilt Chat Model schema that converts messages and tools into industry standard openai formats that can be used downstream with any model for testing. You can also customize the template settings to match your use case.Please see the dataset transformations reference for more information.

​
Create and manage dataset splits

For an overview of when and why to use splits, refer to Dataset organization. To create and manage splits in the UI:
  1. Select examples in your dataset.
  2. Click Add to Split.
  3. From the resulting popup menu, you can select and unselect splits for the selected examples, or create a new split.
Add to Split

​
Edit example metadata

To add metadata to your examples:
  1. Click on an example and then click Edit on the top right-hand side of the popover.
  2. From this page, update or delete existing metadata, or add new metadata.
You may use this to store information about your examples, such as tags or version info, which you can then group by when analyzing experiment results or filter by when you call list_examples in the SDK.

​
Filter examples

You can filter examples by split, metadata key/value or perform full-text search over examples. These filtering options are available to the top left of the examples table:
  • Filter by split: Select split > Select a split to filter by.
  • Filter by metadata: Filters > Select Metadata from the dropdown > Select the metadata key and value to filter on.
  • Full-text search: Filters > Select Full Text from the dropdown > Enter your search criteria.
You may add multiple filters, and only examples that satisfy all of the filters will be displayed in the table. Filters Applied to Examples
Edit this page on GitHub or file an issue.
Connect these docs to Claude, VSCode, and more via MCP for real-time answers.