Run started 2026-10-18 11:02 UTC
· completed 11:13 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
83.90s
p50
82.10s
p90
113.88s
p99
135.48s
min
53.90s
max
137.88s
std
25.52s
First Insight Time
Time to the agent's first update, such as a tool call or a reasoning step.
mean
8.93s
p50
7.00s
p90
12.67s
p99
28.09s
min
3.41s
max
29.80s
std
7.64s
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
96.34s
p50
91.38s
p90
122.72s
p99
141.73s
min
60.70s
max
143.84s
std
24.42s
LLM Time
Total time inside LLM spans, summed across every LLM call in the case.
mean
58.03s
p50
57.09s
p90
72.75s
p99
80.96s
min
37.87s
max
81.87s
std
13.42s
Tool Time
Total time executing tools — database queries, API calls and the like.
mean
31.31s
p50
30.12s
p90
50.87s
p99
59.06s
min
15.31s
max
59.97s
std
14.21s
Other Time
Agent time outside the LLM and tools: orchestration overhead.
mean
0.35s
p50
0.00s
p90
1.16s
p99
2.37s
min
0.00s
max
2.50s
std
0.82s
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
68.7%
p50
69.0%
p90
78.4%
p99
83.0%
min
53.1%
max
83.5%
std
9.3%
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
35.8%
p50
38.0%
p90
44.9%
p99
46.6%
min
21.0%
max
46.8%
std
9.7%
Tool Calls
Number of tool spans the agent produced for the case.
mean
11.90
p50
12.00
p90
14.00
p99
14.00
min
8.00
max
14.00
std
2.02
sum
119.00
Total Tokens
LLM tokens consumed across every call in the case (prompt plus completion).