AI Reliability Console
Monitor behavior, manage risk, and ship confidently
Datasets
Evaluation Datasets
Find and manage datasets
Support Quality Eval
Customer support dataset for refund and cancellation flows.
Single-turntext to text1,238,441 rowsUpdated 2/6/2026
supportrefundsprod
Concierge Chat (Multi)
Multi-turn concierge assistant conversations.
Multi-turntext to text5,642,210 rowsUpdated 2/5/2026
chatconciergegolden
Product FAQ v2
FAQ responses grounded with policy docs.
Single-turntext to text284,012 rowsUpdated 2/3/2026
faqpolicy
Call Center Audio QA
Audio call reviews with transcript and agent response.
Single-turnaudio to text198,420 rowsUpdated 2/4/2026
audioqacompliance