Skip to main content
PATCH
Python
Renaming a dataset keeps its slug, so the old slug still resolves. A new name whose slug another dataset in the project already holds is refused with 409.
New columnTypes re-key every record. To rename a column in the same request that deletes another, send columnRenames with { from, to } for each renamed column, so each value stays with the column it came from. Without it, columns left over after matching names pair by position.

Authorizations

X-Auth-Token
string
header
required

Project API key for sending traces and accessing project-scoped resources. Format: sk-lw-... (no underscore). Obtain one by creating a project via the Admin API or the LangWatch UI.

Path Parameters

slugOrId
string
required

Body

application/json
name
string
Minimum string length: 1
columnTypes
object[]
columnRenames
object[]

Which existing column each renamed column came from, as { from, to } names. Sent, values move by these renames and by unchanged names only. Omitted, columns left over after matching names pair by position.

Response

200 - application/json

Success

id
string
required
name
string
required
slug
string
required
columnTypes
object[]
required
createdAt
string<date-time>
required
updatedAt
string<date-time>
required
platformUrl
string
required
Last modified on November 4, 2026