Run started 2026-10-18 07:39 UTC
· completed 07:46 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
73.69s
p50
63.50s
p90
103.45s
p99
165.22s
min
26.06s
max
174.11s
std
36.35s
First Insight Time
Time to the agent's first update, such as a tool call or a reasoning step.
mean
10.08s
p50
9.68s
p90
12.90s
p99
23.50s
min
4.02s
max
25.01s
std
4.93s
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
81.73s
p50
79.25s
p90
117.46s
p99
177.30s
min
3.92s
max
188.96s
std
40.36s
LLM Time
Total time inside LLM spans, summed across every LLM call in the case.
mean
50.90s
p50
46.17s
p90
76.24s
p99
125.06s
min
6.30s
max
131.31s
std
28.85s
Tool Time
Total time executing tools — database queries, API calls and the like.
mean
24.25s
p50
19.30s
p90
46.20s
p99
57.01s
min
0.00s
max
58.46s
std
16.21s
Other Time
Agent time outside the LLM and tools: orchestration overhead.
mean
3.08s
p50
0.73s
p90
6.07s
p99
19.43s
min
0.00s
max
21.41s
std
5.74s
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
64.5%
p50
71.0%
p90
81.4%
p99
83.1%
min
22.7%
max
83.2%
std
17.7%
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
29.5%
p50
27.5%
p90
44.7%
p99
53.2%
min
0.0%
max
54.3%
std
13.8%
Tool Calls
Number of tool spans the agent produced for the case.
mean
10.07
p50
10.00
p90
14.70
p99
15.00
min
1.00
max
15.00
std
3.58
sum
141.00
Total Tokens
LLM tokens consumed across every call in the case (prompt plus completion).