Run started 2026-10-18 08:05 UTC
· completed 08:09 UTC
Headline Metrics
A case passes when (Correctness or String in Response) and Relevance and Tools pass.
Scorers that had nothing to judge are ignored; cases where no scorer produced a result are
counted as no data and excluded from every pass rate.
First Token Time
Time to the first response token (TTFT) — how long before the user sees anything.
mean
75.28s
p50
75.28s
p90
84.44s
p99
86.50s
min
63.84s
max
86.73s
std
16.19s
First Insight Time
Time to the agent's first update, such as a tool call or a reasoning step.
mean
12.22s
p50
12.22s
p90
13.13s
p99
13.34s
min
11.09s
max
13.36s
std
1.61s
Case Duration
Wall-clock time the eval harness measured for the whole case, including scoring and gRPC overhead. Always at least the agent's own time_total.
mean
65.95s
p50
78.92s
p90
117.36s
p99
125.95s
min
7.90s
max
126.91s
std
44.88s
LLM Time
Total time inside LLM spans, summed across every LLM call in the case.
mean
52.34s
p50
52.34s
p90
57.89s
p99
59.13s
min
45.42s
max
59.27s
std
9.79s
Tool Time
Total time executing tools — database queries, API calls and the like.
mean
25.30s
p50
25.30s
p90
25.45s
p99
25.49s
min
25.12s
max
25.49s
std
0.26s
Other Time
Agent time outside the LLM and tools: orchestration overhead.
mean
3.72s
p50
3.72s
p90
6.70s
p99
7.37s
min
0.00s
max
7.44s
std
5.26s
LLM Time Share
LLM time as a percentage of the agent's total span duration. This is the spread of per-case percentages, not the run-level ratio.
mean
66.5%
p50
66.5%
p90
68.1%
p99
68.5%
min
64.5%
max
68.5%
std
2.8%
Tool Time Share
Tool time as a percentage of the agent's total span duration. This is the spread of per-case percentages, not the run-level ratio.
mean
32.9%
p50
32.9%
p90
37.3%
p99
38.3%
min
27.4%
max
38.5%
std
7.8%
Tool Calls
Number of tool spans the agent produced for the case.
mean
15.00
p50
15.00
p90
16.60
p99
16.96
min
13.00
max
17.00
std
2.83
sum
30.00
Total Tokens
LLM tokens consumed across every call in the case (prompt plus completion).