Reintentar una ejecución de evaluación
Las colas de una evaluación fallida o cancelada se reprograman para reintentar en su lugar. Las muestras completadas conservan sus resultados, las muestras canceladas generadas pero no puntuadas se reanudan en la puntuación, las muestras inacabadas se vuelven a encolar, y las ejecuciones sin muestras persistidas se preparan desde cero con la misma ID de ejecución.
Autorizaciones
Gateway auth: send Authorization: Bearer <mka1-api-key>. For multi-user server-side integrations, you can also send X-On-Behalf-Of: <external-user-id>.
Encabezados
Optional external end-user identifier forwarded by the API gateway.
Parámetros de ruta
Respuesta
Está bien
-9007199254740991 <= x <= 9007199254740991The org that owns this run.
The team that owns this run.
queued, in_progress, finalizing, completed, failed, cancelling, cancelled Coding-agent harness that produced this run (e.g. omp, claude-code). Null for every non-harness eval kind — a null here is not a missing value, it means the run is not an agent run.
Version of agent_name, as the harness reported it. Null when agent_name is.
Reasoning-effort setting the agent ran at (e.g. medium, xhigh). Part of the leaderboard's row identity: the same agent and model at two efforts are two rows, not one. Null when agent_name is.
Total spend for the run in USD. Null when the harness did not report cost.
-9007199254740991 <= x <= 9007199254740991-9007199254740991 <= x <= 9007199254740991-9007199254740991 <= x <= 9007199254740991-9007199254740991 <= x <= 9007199254740991-9007199254740991 <= x <= 9007199254740991