Run started 2026-10-18 08:15 UTC
· completed 08:21 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
115.61s
p50
98.97s
p90
211.29s
p99
216.28s
min
31.51s
max
216.84s
std
59.37s
First Insight Time
Time to the agent's first update, such as a tool call or a reasoning step.
mean
10.24s
p50
9.99s
p90
14.29s
p99
16.23s
min
4.32s
max
16.44s
std
3.91s
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
131.24s
p50
130.66s
p90
223.67s
p99
225.59s
min
40.23s
max
225.80s
std
59.09s
LLM Time
Total time inside LLM spans, summed across every LLM call in the case.
mean
81.78s
p50
68.66s
p90
125.58s
p99
158.75s
min
18.70s
max
162.44s
std
42.26s
Tool Time
Total time executing tools — database queries, API calls and the like.
mean
35.56s
p50
29.52s
p90
46.93s
p99
97.69s
min
10.12s
max
103.33s
std
24.88s
Other Time
Agent time outside the LLM and tools: orchestration overhead.
mean
2.22s
p50
2.24s
p90
5.36s
p99
5.96s
min
0.00s
max
6.03s
std
2.19s
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
69.1%
p50
70.8%
p90
78.0%
p99
86.6%
min
49.6%
max
87.6%
std
11.3%
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
30.8%
p50
27.9%
p90
47.6%
p99
53.6%
min
17.2%
max
54.2%
std
12.0%
Tool Calls
Number of tool spans the agent produced for the case.
mean
14.36
p50
16.00
p90
16.00
p99
20.50
min
10.00
max
21.00
std
3.23
sum
158.00
Total Tokens
LLM tokens consumed across every call in the case (prompt plus completion).