Test summary

Prompt
48 Tools Run v28
Dataset
mcp-toolcall-acc
Status
15 Completed
Test run duration
2 minutes 38 seconds
Evaluation cost
-
Started At.
Jul 4, 2025, 10:44:52 AM
Author
MMadhu Shantan

Summary by evaluator

Tool Call Accuracy
ResultMean scorePass rate
Fail
0.5353.33%

Tokens

TypeValue
Total tokens729025
Input tokens675874
Completion tokens10420

Cost

TypeValue
Total cost$ 0.8708924999999998
Input token cost$ 0.8448424999999999
Completion token cost$ 0.02605

Latency (ms)

TypeValue
min11208.00 ms
max137130.79 ms
p5028607.00 ms
p90 98943.00 ms
p95 137215.00 ms
p99 137215.00 ms
mean 43461.07 ms
stdDeviation36990.0823
total15