eval_run_idstring uuid required
Unique identifier for this run
agent_version_idstring uuid required
Agent version being evaluated
validator_hotkeystring required
Validator performing evaluation
status'CLAIMED' | 'RUNNING' | 'SUCCESS' | 'FAILED' | 'TIMED_OUT' | 'STALE' | 'CANCELLED' required
Status of an evaluation run through its lifecycle.
claimed_atstring date-time required
lease_expires_atstring date-time required
last_heartbeat_atstring date-time nullable
completed_atstring date-time nullable
is_includedboolean required
Whether run is included in aggregate
invalidated_atstring date-time nullable
invalidated_bystring nullable
invalidation_reasonstring nullable
failure_reasonstring nullable
Validator-reported failure reason
validator_scorenumber nullable
reasoning_qualitynumber nullable
Average reasoning quality from LLM judge (0-1)
reasoning_coefficientnumber nullable
Reasoning coefficient applied to outcome score (0.3-1.0)
score_components_summaryobject nullable
Score breakdown on success
service_versionsobject nullable
Snapshot of validator service versions at claim time
sandbox_metadataobject nullable
Sandbox execution details
inference_failure_countinteger nullable
Number of failed miner inference calls
inference_totalinteger nullable
Total number of miner inference calls
phasestring nullable
Evaluation phase (QUALIFYING or RACE)
race_idstring uuid nullable
Race ID if this is a race-phase evaluation
race_numberinteger nullable
Race number if this is a race-phase evaluation