Skip to main content
POST
Dry-run a custom-function evaluation task

Authorizations

x-api-key
string
header
required

Headers

x-selected-account-id
string | null

Body

application/json
task_type
enum<string>
required

The type of evaluation task to dry-run.

Available options:
chat_completion,
inference,
application_variant,
agentex_output,
metric,
auto_evaluation.question,
auto_evaluation.guided_decoding,
auto_evaluation.agent,
contributor_evaluation.question,
contributor_audit,
custom_function
configuration
Configuration · object
required

Task-type-specific configuration. Validated per task_type.

sample_data
Sample Data · object[]
required

Sample data rows to test the task against.

Required array length: 1 - 10 elements

Response

Successful Response

task_type
enum<string>
required
Available options:
chat_completion,
inference,
application_variant,
agentex_output,
metric,
auto_evaluation.question,
auto_evaluation.guided_decoding,
auto_evaluation.agent,
contributor_evaluation.question,
contributor_audit,
custom_function
results
TaskDryRunRowResult · object[]
required
info
Info · object

Task-type-specific metadata (e.g. function_name, parameters).