Run started 2026-10-18 21:59 UTC
· completed 22:08 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
87.50s
p50
80.23s
p90
131.47s
p99
143.12s
min
43.72s
max
144.41s
std
31.41s
First Insight Time
Time to the agent's first update, such as a tool call or a reasoning step.
mean
15.14s
p50
5.83s
p90
28.05s
p99
73.84s
min
4.96s
max
78.93s
std
24.14s
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
102.38s
p50
90.68s
p90
142.83s
p99
165.62s
min
69.13s
max
168.15s
std
31.29s
LLM Time
Total time inside LLM spans, summed across every LLM call in the case.
mean
60.70s
p50
48.79s
p90
106.54s
p99
118.58s
min
29.54s
max
119.92s
std
31.48s
Tool Time
Total time executing tools — database queries, API calls and the like.
mean
32.39s
p50
31.51s
p90
48.55s
p99
55.31s
min
15.94s
max
56.06s
std
13.62s
Other Time
Agent time outside the LLM and tools: orchestration overhead.
mean
0.24s
p50
0.00s
p90
0.43s
p99
1.96s
min
0.00s
max
2.13s
std
0.71s
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
65.5%
p50
73.7%
p90
79.3%
p99
79.5%
min
41.3%
max
79.5%
std
16.0%
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
37.7%
p50
38.2%
p90
58.4%
p99
59.3%
min
20.6%
max
59.4%
std
15.6%
Tool Calls
Number of tool spans the agent produced for the case.
mean
9.78
p50
10.00
p90
11.60
p99
13.76
min
7.00
max
14.00
std
2.11
sum
88.00
Total Tokens
LLM tokens consumed across every call in the case (prompt plus completion).