Run started 2026-10-18 10:46 UTC
· completed 11:02 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
87.61s
p50
82.75s
p90
123.37s
p99
129.35s
min
49.52s
max
130.02s
std
26.01s
First Insight Time
Time to the agent's first update, such as a tool call or a reasoning step.
mean
7.01s
p50
6.60s
p90
10.21s
p99
11.01s
min
3.97s
max
11.10s
std
2.38s
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
111.14s
p50
91.71s
p90
137.69s
p99
249.98s
min
59.65s
max
262.46s
std
55.72s
LLM Time
Total time inside LLM spans, summed across every LLM call in the case.
mean
58.70s
p50
55.84s
p90
77.34s
p99
84.40s
min
33.04s
max
85.18s
std
15.98s
Tool Time
Total time executing tools — database queries, API calls and the like.
mean
36.06s
p50
35.28s
p90
50.58s
p99
73.73s
min
12.15s
max
76.30s
std
19.04s
Other Time
Agent time outside the LLM and tools: orchestration overhead.
mean
0.18s
p50
0.00s
p90
0.20s
p99
1.61s
min
0.00s
max
1.77s
std
0.56s
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
66.6%
p50
66.2%
p90
80.5%
p99
83.9%
min
46.2%
max
84.3%
std
12.4%
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
39.0%
p50
38.6%
p90
55.2%
p99
57.1%
min
15.7%
max
57.3%
std
13.8%
Tool Calls
Number of tool spans the agent produced for the case.
mean
13.50
p50
13.50
p90
18.00
p99
18.00
min
7.00
max
18.00
std
3.69
sum
135.00
Total Tokens
LLM tokens consumed across every call in the case (prompt plus completion).