Eval scorecard

Win rate
92.4%
+4.1%
Hallucination
1,248
-3.2%
Latency
37
watch
Eval scorecard
Line chart visualization
Recent Evaluations
Evaluation ID Win Rate Hallucinations Latency (ms)
EVAL-2023-04 92.4% 1,248 37
EVAL-2023-03 89.7% 1,452 42
EVAL-2023-02 91.2% 1,301 39