multi-agent-research-assistant / evaluation_report.json
mohmad017's picture
Multi-Agent Research Assistant — LangGraph + FAISS + RAG + Evaluation
4619ed7
Raw
History Blame Contribute Delete
281 Bytes
{
"num_questions": 20,
"metrics": {
"faithfulness": 0.99,
"answer_relevancy": 0.82
},
"latency_ms": {
"p50": 2218.5,
"p90": 2580.9
},
"targets_met": {
"faithfulness_gt_085": true,
"answer_relevancy_gt_080": true,
"p90_lt_2000ms": false
}
}