Skip to main content
POST
Edit a benchmark from a JSON spec

Authorizations

Authorization
string
header
required

Platform API token created under Profile Settings > Platform API tokens. Scopes are fixed at creation.

Path Parameters

id
integer
required

Benchmark id

Query Parameters

dry_run
boolean

Validate only: answers 200 with a summary and a preview, and changes nothing

Body

application/json

At least one of title, description, terms, pages, contact_email, organization_name, reward, report or items

title
string
Example:

"Unit Conversion Basics, revised"

description
string
terms
string
contact_email
string<email>
organization_name
string
Maximum string length: 256
reward
string
Maximum string length: 256
report
string
Maximum string length: 256
pages
object[]

The complete new page list, in display order, replacing the old one

Maximum array length: 20
type
enum<string>

Optional and only together with items; inferred from the items when omitted

Available options:
question_answer,
multiple_choice
items
object[]

The complete new question list, replacing the old one

Required array length: 5 - 2000 elements

Response

Edited, or with dry_run the summary of what would change

valid
boolean
id
integer
title
string
slug
string
published
boolean
changed
string[]

Which of title, description and items actually changed

kind
string

Present when items were sent

pages
integer | null

How many pages were sent, null when the edit left them alone

items
integer | null

How many items were sent, null when the edit left the questions alone

tasks_updated
integer

How many tasks got the rebuilt datasets

preview
object[]

dry_run only: the first few items as the platform read them

Last modified on September 16, 2026