HTTP API
EvalRun
A single execution record and summary of an EvalSuite.
Shape: object
| Property | Required | Type | Description | Rules |
|---|---|---|---|---|
id |
yes | string (uuid) |
The run identifier. | — |
tenantId |
yes | string |
The tenant the run belongs to. | — |
suiteId |
yes | string (uuid) |
The identifier of the suite being measured. | — |
jobId |
no | string (uuid) |
The identifier of the job record responsible for executing this run. The run is executed through the job queue (IJobStore). | — |
agentVersion |
no | integer (int32) |
The agent definition version being measured. null before the run starts. |
pattern `^-?(?:0\ |
modelId |
no | string |
The model identifier being measured. null before the run starts. |
— |
status |
yes | EvalRunStatus |
The current status of the run. | — |
total |
no | integer (int32) |
The total number of cases. | pattern `^-?(?:0\ |
passed |
no | integer (int32) |
The number of cases that passed. | pattern `^-?(?:0\ |
failed |
no | integer (int32) |
The number of remaining (failed) cases. | pattern `^-?(?:0\ |
inputTokens |
no | integer (int64) |
The total input token count. | pattern `^-?(?:0\ |
outputTokens |
no | integer (int64) |
The total output token count. | pattern `^-?(?:0\ |
startedAt |
yes | string (date-time) |
The start time (UTC). | — |
completedAt |
no | string (date-time) |
The completion time (UTC). null while the run is in progress. |
— |