Run started 2026-10-18 07:01 UTC
· completed 07:08 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
97.48s
p50
79.52s
p90
146.16s
p99
176.81s
min
43.48s
max
180.16s
std
37.98s
First Insight Time
Time to the agent's first update, such as a tool call or a reasoning step.
mean
9.45s
p50
9.00s
p90
14.40s
p99
16.04s
min
4.61s
max
16.20s
std
3.77s
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
109.76s
p50
87.90s
p90
154.77s
p99
190.52s
min
54.38s
max
194.90s
std
38.46s
LLM Time
Total time inside LLM spans, summed across every LLM call in the case.
mean
70.87s
p50
59.00s
p90
114.95s
p99
134.91s
min
37.24s
max
137.20s
std
30.85s
Tool Time
Total time executing tools — database queries, API calls and the like.
mean
30.92s
p50
30.40s
p90
46.78s
p99
51.10s
min
7.59s
max
51.43s
std
11.54s
Other Time
Agent time outside the LLM and tools: orchestration overhead.
mean
1.16s
p50
0.00s
p90
3.29s
p99
5.93s
min
0.00s
max
6.28s
std
1.89s
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
70.0%
p50
70.8%
p90
80.6%
p99
83.7%
min
54.5%
max
83.7%
std
8.9%
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
31.9%
p50
31.6%
p90
41.2%
p99
53.9%
min
15.6%
max
56.0%
std
10.7%
Tool Calls
Number of tool spans the agent produced for the case.
mean
12.00
p50
12.00
p90
15.00
p99
15.86
min
8.00
max
16.00
std
2.54
sum
180.00
Total Tokens
LLM tokens consumed across every call in the case (prompt plus completion).