Run: run_letta_pm_001
VERIFIED
SystemLetta
BenchmarkPersonaMem
Harnessv0.9.4
Verified Overall87.1%
Nuance Overall81.2%
DateMay 7, 2026
run_manifest.jsonjson
{
"version": "1.0.0",
"runId": "run_letta_pm_001",
"systemId": "sys_letta",
"systemName": "Letta",
"benchmarkId": "bench_personamem",
"benchmarkName": "PersonaMem",
"benchmarkVersion": "2.0",
"harnessVersion": "0.9.4",
"judgeModel": "gpt-4o-2025-03-26",
"judgeTemperature": 0,
"startedAt": "2026-05-07T14:22:00Z",
"completedAt": "2026-05-07T16:05:38Z",
"scores": {
"verified": {
"recall": 89.2,
"temporal": 83.7,
"reasoning": 88.4,
"overall": 87.1
},
"nuance": {
"recall": 83.4,
"temporal": 77.1,
"reasoning": 82.9,
"overall": 81.2
}
},
"questionCount": 200,
"passCount": 171,
"failCount": 29,
"merkleRoot": "d1e2f3a4b5c6d7e8f9a0b1c2d3e4f5a6b7c8d9e0f1a2b3c4d5e6f7a8b9c0d1e2"
}