Run started 2026-10-17 09:43 UTC
· completed 09:52 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
83.91s
p50
77.75s
p90
115.90s
p99
123.55s
min
49.58s
max
124.40s
std
26.09s
First Insight Time
Time to the agent's first update, such as a tool call or a reasoning step.
mean
11.36s
p50
10.62s
p90
15.36s
p99
20.87s
min
3.34s
max
21.48s
std
5.02s
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
94.63s
p50
93.81s
p90
133.99s
p99
173.69s
min
33.78s
max
178.10s
std
38.63s
LLM Time
Total time inside LLM spans, summed across every LLM call in the case.
mean
59.71s
p50
58.98s
p90
74.32s
p99
80.25s
min
37.12s
max
80.91s
std
14.45s
Tool Time
Total time executing tools — database queries, API calls and the like.
mean
29.15s
p50
24.25s
p90
50.77s
p99
69.48s
min
4.63s
max
71.56s
std
21.02s
Other Time
Agent time outside the LLM and tools: orchestration overhead.
mean
1.06s
p50
0.00s
p90
3.79s
p99
5.24s
min
0.00s
max
5.40s
std
2.06s
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
71.7%
p50
74.7%
p90
84.6%
p99
85.2%
min
39.9%
max
85.2%
std
14.5%
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
31.7%
p50
31.2%
p90
48.3%
p99
60.0%
min
6.8%
max
61.3%
std
16.5%
Tool Calls
Number of tool spans the agent produced for the case.
mean
11.62
p50
11.50
p90
15.60
p99
16.86
min
8.00
max
17.00
std
3.16
sum
93.00
Total Tokens
LLM tokens consumed across every call in the case (prompt plus completion).