Run started 2026-10-18 07:17 UTC
· completed 07:24 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
105.58s
p50
108.45s
p90
157.85s
p99
181.74s
min
51.55s
max
184.40s
std
47.08s
First Insight Time
Time to the agent's first update, such as a tool call or a reasoning step.
mean
10.13s
p50
10.06s
p90
14.37s
p99
15.59s
min
5.60s
max
15.73s
std
3.57s
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
125.49s
p50
117.72s
p90
187.38s
p99
286.57s
min
57.45s
max
300.00s
std
66.77s
LLM Time
Total time inside LLM spans, summed across every LLM call in the case.
mean
78.93s
p50
76.59s
p90
116.30s
p99
135.51s
min
45.26s
max
137.64s
std
31.20s
Tool Time
Total time executing tools — database queries, API calls and the like.
mean
31.17s
p50
24.18s
p90
58.21s
p99
79.62s
min
0.32s
max
82.00s
std
26.18s
Other Time
Agent time outside the LLM and tools: orchestration overhead.
mean
1.54s
p50
0.00s
p90
3.64s
p99
9.17s
min
0.00s
max
9.79s
std
3.21s
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
76.2%
p50
76.2%
p90
91.4%
p99
102.7%
min
42.4%
max
104.0%
std
17.1%
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
26.3%
p50
23.0%
p90
40.6%
p99
64.4%
min
0.4%
max
67.0%
std
18.4%
Tool Calls
Number of tool spans the agent produced for the case.
mean
11.11
p50
10.00
p90
16.40
p99
17.84
min
2.00
max
18.00
std
4.57
sum
100.00
Total Tokens
LLM tokens consumed across every call in the case (prompt plus completion).