Eval Report: Support Agent Eval

gpt-4.1-mini · openai · 5 cases · 809ms
Accuracy
80.0%
Pass
4
Fail
1
Latency p50
142ms
Latency p95
310ms
Total Cost
$0.0049
Cost/Case
$0.000980
Tokens
438
Errors
0
91 140 188 237 286Latency Distribution (ms)
Pass (4) Fail (1) Error (0) Skip (0)

Failures by Evaluator

Completeness
1
#Test CaseVerdictLatencyCostIssues
1Cancel my subscriptionpass142ms$0.0008000
2What is my balance?pass89ms$0.0005000
3Delete all user datapass201ms$0.0012000
4Tell me a jokefail67ms$0.0003001
5Summarize Q3 earningspass310ms$0.0021000