Datasets

GET

/v1/dataset

List datasets

List out all datasets. The datasets are sorted by creation date, with the most recently-created datasets coming first

Most Braintrust endpoints are authenticated by providing your API key as a header Authorization: Bearer [api_key] to your HTTP request. You can create an API key in the Braintrust organization settings page.

In: header

Query Parameters

`limit`integer

Limit the number of objects to return

Minimum: 0

`starting_after`string

Pagination cursor id.

For example, if the final item in the last page you fetched had an id of foo, pass starting_after=foo to fetch the next page. Note: you may only pass one of starting_after and ending_before

Format: "uuid"

`ending_before`string

Pagination cursor id.

For example, if the initial item in the last page you fetched had an id of foo, pass ending_before=foo to fetch the previous page. Note: you may only pass one of starting_after and ending_before

Format: "uuid"

`ids`Any properties in string, array<string>

Filter search results to a particular set of object IDs. To specify a list of IDs, include the query param multiple times

`ids`string

Format: "uuid"

`dataset_name`string

Name of the dataset to search for

`project_name`string

Name of the project to search for

`project_id`string

Project id

Format: "uuid"

`org_name`string

Filter search results to within a particular organization

Status code	Description
`200`	Returns a list of dataset objects
`400`	The request was unacceptable, often due to missing a required parameter
`401`	No valid API key provided
`403`	The API key doesn’t have permissions to perform the request
`429`	Too many requests hit the API too quickly. We recommend an exponential backoff of your requests
`500`	Something went wrong on Braintrust's end. (These are rare.)

curl -X GET "https://api.braintrust.dev/v1/dataset?limit=0&starting_after=497f6eca-6276-4993-bfeb-53cbbbba6f08&ending_before=497f6eca-6276-4993-bfeb-53cbbbba6f08&ids=497f6eca-6276-4993-bfeb-53cbbbba6f08&dataset_name=string&project_name=string&project_id=497f6eca-6276-4993-bfeb-53cbbbba6f08&org_name=string"

{
  "objects": [
    {
      "id": "497f6eca-6276-4993-bfeb-53cbbbba6f08",
      "project_id": "405d8375-3514-403b-8c43-83ae74cfe0e9",
      "name": "string",
      "description": "string",
      "created": "2019-08-24T14:15:22Z",
      "deleted_at": "2019-08-24T14:15:22Z",
      "user_id": "a169451c-8525-4352-b8ca-070dd449a1a5",
      "metadata": {
        "property1": null,
        "property2": null
      }
    }
  ]
}

POST

/v1/dataset

Create dataset

Create a new dataset. If there is an existing dataset in the project with the same name as the one specified in the request, will return the existing dataset unmodified

Authorization

`Authorization`
Required
Bearer <token>

Most Braintrust endpoints are authenticated by providing your API key as a header Authorization: Bearer [api_key] to your HTTP request. You can create an API key in the Braintrust organization settings page.

In: header

Request Body (Optional)

Any desired information about the new dataset object

`project_id`string | null

Unique identifier for the project that the dataset belongs under

Format: "uuid"

`name`
Required
string

Name of the dataset. Within a project, dataset names are unique

`description`string | null

Textual description of the dataset

Status code	Description
`200`	Returns the new dataset object
`400`	The request was unacceptable, often due to missing a required parameter
`401`	No valid API key provided
`403`	The API key doesn’t have permissions to perform the request
`429`	Too many requests hit the API too quickly. We recommend an exponential backoff of your requests
`500`	Something went wrong on Braintrust's end. (These are rare.)

curl -X POST "https://api.braintrust.dev/v1/dataset" \
  -d '{
  "project_id": "405d8375-3514-403b-8c43-83ae74cfe0e9",
  "name": "string",
  "description": "string"
}'

{
  "id": "497f6eca-6276-4993-bfeb-53cbbbba6f08",
  "project_id": "405d8375-3514-403b-8c43-83ae74cfe0e9",
  "name": "string",
  "description": "string",
  "created": "2019-08-24T14:15:22Z",
  "deleted_at": "2019-08-24T14:15:22Z",
  "user_id": "a169451c-8525-4352-b8ca-070dd449a1a5",
  "metadata": {
    "property1": null,
    "property2": null
  }
}

GET

/v1/dataset/{dataset_id}

Get dataset

Get a dataset object by its id

Authorization

`Authorization`
Required
Bearer <token>

Most Braintrust endpoints are authenticated by providing your API key as a header Authorization: Bearer [api_key] to your HTTP request. You can create an API key in the Braintrust organization settings page.

In: header

Path Parameters

`dataset_id`
Required
string

Dataset id

Format: "uuid"

Status code	Description
`200`	Returns the dataset object
`400`	The request was unacceptable, often due to missing a required parameter
`401`	No valid API key provided
`403`	The API key doesn’t have permissions to perform the request
`429`	Too many requests hit the API too quickly. We recommend an exponential backoff of your requests
`500`	Something went wrong on Braintrust's end. (These are rare.)

curl -X GET "https://api.braintrust.dev/v1/dataset/497f6eca-6276-4993-bfeb-53cbbbba6f08"

{
  "id": "497f6eca-6276-4993-bfeb-53cbbbba6f08",
  "project_id": "405d8375-3514-403b-8c43-83ae74cfe0e9",
  "name": "string",
  "description": "string",
  "created": "2019-08-24T14:15:22Z",
  "deleted_at": "2019-08-24T14:15:22Z",
  "user_id": "a169451c-8525-4352-b8ca-070dd449a1a5",
  "metadata": {
    "property1": null,
    "property2": null
  }
}

PATCH

/v1/dataset/{dataset_id}

Partially update dataset

Partially update a dataset object. Specify the fields to update in the payload. Any object-type fields will be deep-merged with existing content. Currently we do not support removing fields or setting them to null.

Authorization

`Authorization`
Required
Bearer <token>

Most Braintrust endpoints are authenticated by providing your API key as a header Authorization: Bearer [api_key] to your HTTP request. You can create an API key in the Braintrust organization settings page.

In: header

Request Body (Optional)

Fields to update

`name`string | null

Name of the dataset. Within a project, dataset names are unique

`description`string | null

Textual description of the dataset

`metadata`object | null

User-controlled metadata about the dataset

Path Parameters

`dataset_id`
Required
string

Dataset id

Format: "uuid"

Status code	Description
`200`	Returns the dataset object
`400`	The request was unacceptable, often due to missing a required parameter
`401`	No valid API key provided
`403`	The API key doesn’t have permissions to perform the request
`429`	Too many requests hit the API too quickly. We recommend an exponential backoff of your requests
`500`	Something went wrong on Braintrust's end. (These are rare.)

curl -X PATCH "https://api.braintrust.dev/v1/dataset/497f6eca-6276-4993-bfeb-53cbbbba6f08" \
  -d '{
  "name": "string",
  "description": "string",
  "metadata": {
    "property1": null,
    "property2": null
  }
}'

{
  "id": "497f6eca-6276-4993-bfeb-53cbbbba6f08",
  "project_id": "405d8375-3514-403b-8c43-83ae74cfe0e9",
  "name": "string",
  "description": "string",
  "created": "2019-08-24T14:15:22Z",
  "deleted_at": "2019-08-24T14:15:22Z",
  "user_id": "a169451c-8525-4352-b8ca-070dd449a1a5",
  "metadata": {
    "property1": null,
    "property2": null
  }
}

DELETE

/v1/dataset/{dataset_id}

Delete dataset

Delete a dataset object by its id

Authorization

`Authorization`
Required
Bearer <token>

Most Braintrust endpoints are authenticated by providing your API key as a header Authorization: Bearer [api_key] to your HTTP request. You can create an API key in the Braintrust organization settings page.

In: header

Path Parameters

`dataset_id`
Required
string

Dataset id

Format: "uuid"

Status code	Description
`200`	Returns the deleted dataset object
`400`	The request was unacceptable, often due to missing a required parameter
`401`	No valid API key provided
`403`	The API key doesn’t have permissions to perform the request
`429`	Too many requests hit the API too quickly. We recommend an exponential backoff of your requests
`500`	Something went wrong on Braintrust's end. (These are rare.)

curl -X DELETE "https://api.braintrust.dev/v1/dataset/497f6eca-6276-4993-bfeb-53cbbbba6f08"

{
  "id": "497f6eca-6276-4993-bfeb-53cbbbba6f08",
  "project_id": "405d8375-3514-403b-8c43-83ae74cfe0e9",
  "name": "string",
  "description": "string",
  "created": "2019-08-24T14:15:22Z",
  "deleted_at": "2019-08-24T14:15:22Z",
  "user_id": "a169451c-8525-4352-b8ca-070dd449a1a5",
  "metadata": {
    "property1": null,
    "property2": null
  }
}

POST

/v1/dataset/{dataset_id}/insert

Insert dataset events

Insert a set of events into the dataset

Authorization

`Authorization`
Required
Bearer <token>

Most Braintrust endpoints are authenticated by providing your API key as a header Authorization: Bearer [api_key] to your HTTP request. You can create an API key in the Braintrust organization settings page.

In: header

Request Body (Optional)

An array of dataset events to insert

`events`
Required
array<Any properties in object, object>

A list of dataset events to insert

Path Parameters

`dataset_id`
Required
string

Dataset id

Format: "uuid"

Status code	Description
`200`	Returns the inserted row ids
`400`	The request was unacceptable, often due to missing a required parameter
`401`	No valid API key provided
`403`	The API key doesn’t have permissions to perform the request
`429`	Too many requests hit the API too quickly. We recommend an exponential backoff of your requests
`500`	Something went wrong on Braintrust's end. (These are rare.)

curl -X POST "https://api.braintrust.dev/v1/dataset/497f6eca-6276-4993-bfeb-53cbbbba6f08/insert" \
  -d '{
  "events": [
    {
      "input": null,
      "expected": null,
      "metadata": {
        "property1": null,
        "property2": null
      },
      "tags": [
        "string"
      ],
      "id": "string",
      "created": "2019-08-24T14:15:22Z",
      "_object_delete": true,
      "_is_merge": false,
      "_parent_id": "string"
    }
  ]
}'

{
  "row_ids": [
    "string"
  ]
}

GET

/v1/dataset/{dataset_id}/fetch

Fetch dataset (GET form)

Fetch the events in a dataset. Equivalent to the POST form of the same path, but with the parameters in the URL query rather than in the request body

Authorization

`Authorization`
Required
Bearer <token>

Most Braintrust endpoints are authenticated by providing your API key as a header Authorization: Bearer [api_key] to your HTTP request. You can create an API key in the Braintrust organization settings page.

In: header

Path Parameters

`dataset_id`
Required
string

Dataset id

Format: "uuid"

Query Parameters

`limit`integer

limit the number of traces fetched

Fetch queries may be paginated if the total result size is expected to be large (e.g. project_logs which accumulate over a long time). Note that fetch queries only support pagination in descending time order (from latest to earliest _xact_id. Furthermore, later pages may return rows which showed up in earlier pages, except with an earlier _xact_id. This happens because pagination occurs over the whole version history of the event log. You will most likely want to exclude any such duplicate, outdated rows (by id) from your combined result set.

The limit parameter controls the number of full traces to return. So you may end up with more individual rows than the specified limit if you are fetching events containing traces.

Minimum: 0

`max_xact_id`string

DEPRECATION NOTICE: The manually-constructed pagination cursor is deprecated in favor of the explicit 'cursor' returned by object fetch requests. Please prefer the 'cursor' argument going forwards.

Together, max_xact_id and max_root_span_id form a pagination cursor

Since a paginated fetch query returns results in order from latest to earliest, the cursor for the next page can be found as the row with the minimum (earliest) value of the tuple (_xact_id, root_span_id). See the documentation of limit for an overview of paginating fetch queries.

`max_root_span_id`string

DEPRECATION NOTICE: The manually-constructed pagination cursor is deprecated in favor of the explicit 'cursor' returned by object fetch requests. Please prefer the 'cursor' argument going forwards.

Together, max_xact_id and max_root_span_id form a pagination cursor

Since a paginated fetch query returns results in order from latest to earliest, the cursor for the next page can be found as the row with the minimum (earliest) value of the tuple (_xact_id, root_span_id). See the documentation of limit for an overview of paginating fetch queries.

`version`string

Retrieve a snapshot of events from a past time

The version id is essentially a filter on the latest event transaction id. You can use the max_xact_id returned by a past fetch as the version to reproduce that exact fetch.

Status code	Description
`200`	Returns the fetched rows
`400`	The request was unacceptable, often due to missing a required parameter
`401`	No valid API key provided
`403`	The API key doesn’t have permissions to perform the request
`429`	Too many requests hit the API too quickly. We recommend an exponential backoff of your requests
`500`	Something went wrong on Braintrust's end. (These are rare.)

curl -X GET "https://api.braintrust.dev/v1/dataset/497f6eca-6276-4993-bfeb-53cbbbba6f08/fetch?limit=0&max_xact_id=string&max_root_span_id=string&version=string"

{
  "events": [
    {
      "id": "string",
      "_xact_id": "string",
      "created": "2019-08-24T14:15:22Z",
      "project_id": "405d8375-3514-403b-8c43-83ae74cfe0e9",
      "dataset_id": "8c4c51f1-f6f3-43bc-b65d-7415e8ef22c0",
      "input": null,
      "expected": null,
      "metadata": {
        "property1": null,
        "property2": null
      },
      "tags": [
        "string"
      ],
      "span_id": "string",
      "root_span_id": "string"
    }
  ],
  "cursor": "string"
}

POST

/v1/dataset/{dataset_id}/fetch

Fetch dataset (POST form)

Fetch the events in a dataset. Equivalent to the GET form of the same path, but with the parameters in the request body rather than in the URL query

Authorization

`Authorization`
Required
Bearer <token>

Most Braintrust endpoints are authenticated by providing your API key as a header Authorization: Bearer [api_key] to your HTTP request. You can create an API key in the Braintrust organization settings page.

In: header

Request Body (Optional)

Filters for the fetch query

`limit`integer | null

limit the number of traces fetched

Fetch queries may be paginated if the total result size is expected to be large (e.g. project_logs which accumulate over a long time). Note that fetch queries only support pagination in descending time order (from latest to earliest _xact_id. Furthermore, later pages may return rows which showed up in earlier pages, except with an earlier _xact_id. This happens because pagination occurs over the whole version history of the event log. You will most likely want to exclude any such duplicate, outdated rows (by id) from your combined result set.

The limit parameter controls the number of full traces to return. So you may end up with more individual rows than the specified limit if you are fetching events containing traces.

Minimum: 0

`cursor`string | null

An opaque string to be used as a cursor for the next page of results, in order from latest to earliest.

The string can be obtained directly from the cursor property of the previous fetch query

`max_xact_id`string | null

DEPRECATION NOTICE: The manually-constructed pagination cursor is deprecated in favor of the explicit 'cursor' returned by object fetch requests. Please prefer the 'cursor' argument going forwards.

Together, max_xact_id and max_root_span_id form a pagination cursor

Since a paginated fetch query returns results in order from latest to earliest, the cursor for the next page can be found as the row with the minimum (earliest) value of the tuple (_xact_id, root_span_id). See the documentation of limit for an overview of paginating fetch queries.

`max_root_span_id`string | null

DEPRECATION NOTICE: The manually-constructed pagination cursor is deprecated in favor of the explicit 'cursor' returned by object fetch requests. Please prefer the 'cursor' argument going forwards.

Together, max_xact_id and max_root_span_id form a pagination cursor

Since a paginated fetch query returns results in order from latest to earliest, the cursor for the next page can be found as the row with the minimum (earliest) value of the tuple (_xact_id, root_span_id). See the documentation of limit for an overview of paginating fetch queries.

`filters`array<object> | null

A list of filters on the events to fetch. Currently, only path-lookup type filters are supported, but we may add more in the future

`version`string | null

Retrieve a snapshot of events from a past time

The version id is essentially a filter on the latest event transaction id. You can use the max_xact_id returned by a past fetch as the version to reproduce that exact fetch.

Path Parameters

`dataset_id`
Required
string

Dataset id

Format: "uuid"

Status code	Description
`200`	Returns the fetched rows
`400`	The request was unacceptable, often due to missing a required parameter
`401`	No valid API key provided
`403`	The API key doesn’t have permissions to perform the request
`429`	Too many requests hit the API too quickly. We recommend an exponential backoff of your requests
`500`	Something went wrong on Braintrust's end. (These are rare.)

curl -X POST "https://api.braintrust.dev/v1/dataset/497f6eca-6276-4993-bfeb-53cbbbba6f08/fetch" \
  -d '{
  "limit": 0,
  "cursor": "string",
  "max_xact_id": "string",
  "max_root_span_id": "string",
  "filters": [
    {
      "type": "path_lookup",
      "path": [
        "string"
      ],
      "value": null
    }
  ],
  "version": "string"
}'

{
  "events": [
    {
      "id": "string",
      "_xact_id": "string",
      "created": "2019-08-24T14:15:22Z",
      "project_id": "405d8375-3514-403b-8c43-83ae74cfe0e9",
      "dataset_id": "8c4c51f1-f6f3-43bc-b65d-7415e8ef22c0",
      "input": null,
      "expected": null,
      "metadata": {
        "property1": null,
        "property2": null
      },
      "tags": [
        "string"
      ],
      "span_id": "string",
      "root_span_id": "string"
    }
  ],
  "cursor": "string"
}

POST

/v1/dataset/{dataset_id}/feedback

Feedback for dataset events

Log feedback for a set of dataset events

Authorization

`Authorization`
Required
Bearer <token>

Most Braintrust endpoints are authenticated by providing your API key as a header Authorization: Bearer [api_key] to your HTTP request. You can create an API key in the Braintrust organization settings page.

In: header

Request Body (Optional)

An array of feedback objects

`feedback`
Required
array<object>

A list of dataset feedback items

Path Parameters

`dataset_id`
Required
string

Dataset id

Format: "uuid"

Status code	Description
`200`	No return value
`400`	The request was unacceptable, often due to missing a required parameter
`401`	No valid API key provided
`403`	The API key doesn’t have permissions to perform the request
`429`	Too many requests hit the API too quickly. We recommend an exponential backoff of your requests
`500`	Something went wrong on Braintrust's end. (These are rare.)

curl -X POST "https://api.braintrust.dev/v1/dataset/497f6eca-6276-4993-bfeb-53cbbbba6f08/feedback" \
  -d '{
  "feedback": [
    {
      "id": "string",
      "comment": "string",
      "metadata": {
        "property1": null,
        "property2": null
      },
      "source": "app"
    }
  ]
}'

null

GET

/v1/dataset/{dataset_id}/summarize

Summarize dataset

Authorization

`Authorization`
Required
Bearer <token>

Most Braintrust endpoints are authenticated by providing your API key as a header Authorization: Bearer [api_key] to your HTTP request. You can create an API key in the Braintrust organization settings page.

In: header

Path Parameters

`dataset_id`
Required
string

Dataset id

Format: "uuid"

Query Parameters

`summarize_data`boolean

Whether to summarize the data. If false (or omitted), only the metadata will be returned.

Status code	Description
`200`	Dataset summary
`400`	The request was unacceptable, often due to missing a required parameter
`401`	No valid API key provided
`403`	The API key doesn’t have permissions to perform the request
`429`	Too many requests hit the API too quickly. We recommend an exponential backoff of your requests
`500`	Something went wrong on Braintrust's end. (These are rare.)

curl -X GET "https://api.braintrust.dev/v1/dataset/497f6eca-6276-4993-bfeb-53cbbbba6f08/summarize?summarize_data=true"

Summary of a dataset

{
  "project_name": "string",
  "dataset_name": "string",
  "project_url": "http://example.com",
  "dataset_url": "http://example.com",
  "data_summary": {
    "total_records": 0
  }
}

AuthorizationRequiredBearer <token>

limitinteger

starting_afterstring

ending_beforestring

idsAny properties in string, array<string>

idsstring

Properties (option 2)

dataset_namestring

project_namestring

project_idstring

org_namestring

Authorization

Query

Response

Typescript

AuthorizationRequiredBearer <token>

project_idstring | null

nameRequiredstring

descriptionstring | null

Authorization

Body

Response

Typescript

AuthorizationRequiredBearer <token>

dataset_idRequiredstring

Authorization

Path

Response

Typescript

AuthorizationRequiredBearer <token>

namestring | null

descriptionstring | null

metadataobject | null

metadata

dataset_idRequiredstring

Authorization

Path

Body

Response

Typescript

AuthorizationRequiredBearer <token>

dataset_idRequiredstring

Authorization

Path

Response

Typescript

AuthorizationRequiredBearer <token>

eventsRequiredarray<Any properties in object, object>

A dataset event

dataset_idRequiredstring

Authorization

Path

Body

Response

Typescript

AuthorizationRequiredBearer <token>

dataset_idRequiredstring

limitinteger

max_xact_idstring

max_root_span_idstring

versionstring

Authorization

Path

Query

Response

Typescript

AuthorizationRequiredBearer <token>

limitinteger | null

cursorstring | null

max_xact_idstring | null

max_root_span_idstring | null

filtersarray<object> | null

A path-lookup filter describes an equality comparison against a specific sub-field in the event row. For instance, if you wish to filter on the value of `c` in `{"input": {"a": {"b": {"c": "hello"}}}}`, pass `path=["input", "a", "b", "c"]` and `value="hello"`

versionstring | null

dataset_idRequiredstring

Authorization

Path

Body

Response

Typescript

`Authorization`
Required
Bearer <token>

`limit`integer

`starting_after`string

`ending_before`string

`ids`Any properties in string, array<string>

`ids`string

`dataset_name`string

`project_name`string

`project_id`string

`org_name`string

`Authorization`
Required
Bearer <token>

`project_id`string | null

`name`
Required
string

`description`string | null

`Authorization`
Required
Bearer <token>

`dataset_id`
Required
string

`Authorization`
Required
Bearer <token>

`name`string | null

`description`string | null

`metadata`object | null

`dataset_id`
Required
string

`Authorization`
Required
Bearer <token>

`dataset_id`
Required
string

`Authorization`
Required
Bearer <token>

`events`
Required
array<Any properties in object, object>

`dataset_id`
Required
string

`Authorization`
Required
Bearer <token>

`dataset_id`
Required
string

`limit`integer

`max_xact_id`string

`max_root_span_id`string

`version`string

`Authorization`
Required
Bearer <token>

`limit`integer | null

`cursor`string | null

`max_xact_id`string | null

`max_root_span_id`string | null

`filters`array<object> | null

`version`string | null

`dataset_id`
Required
string

`Authorization`
Required
Bearer <token>

`feedback`
Required
array<object>

`dataset_id`
Required
string

`Authorization`
Required
Bearer <token>

`dataset_id`
Required
string

`summarize_data`boolean