Run started 2026-10-18 22:48 UTC
· completed 23:53 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
80.73s
p50
75.27s
p90
121.15s
p99
227.86s
min
21.30s
max
234.62s
std
38.03s
First Insight Time
Time to the agent's first update, such as a tool call or a reasoning step.
mean
8.22s
p50
5.14s
p90
13.92s
p99
53.74s
min
2.62s
max
79.08s
std
11.05s
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
94.62s
p50
88.92s
p90
134.26s
p99
240.55s
min
28.24s
max
253.93s
std
39.66s
LLM Time
Total time inside LLM spans, summed across every LLM call in the case.
mean
59.00s
p50
54.70s
p90
87.54s
p99
160.78s
min
20.03s
max
188.46s
std
28.98s
Tool Time
Total time executing tools — database queries, API calls and the like.
mean
29.65s
p50
27.42s
p90
58.91s
p99
97.86s
min
0.97s
max
127.08s
std
22.00s
Other Time
Agent time outside the LLM and tools: orchestration overhead.
mean
0.36s
p50
0.00s
p90
1.62s
p99
2.83s
min
0.00s
max
3.38s
std
0.78s
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
71.1%
p50
70.7%
p90
91.0%
p99
108.0%
min
39.2%
max
127.9%
std
16.1%
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
34.0%
p50
32.1%
p90
56.2%
p99
82.4%
min
4.0%
max
85.0%
std
17.5%
Tool Calls
Number of tool spans the agent produced for the case.
mean
10.81
p50
10.00
p90
15.00
p99
16.32
min
6.00
max
17.00
std
2.73
sum
746.00
Total Tokens
LLM tokens consumed across every call in the case (prompt plus completion).