Run started 2026-10-17 09:00 UTC
· completed 09:08 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
84.19s
p50
73.47s
p90
135.29s
p99
181.91s
min
43.78s
max
184.45s
std
39.76s
First Insight Time
Time to the agent's first update, such as a tool call or a reasoning step.
mean
9.93s
p50
8.29s
p90
16.97s
p99
22.86s
min
3.80s
max
23.79s
std
5.22s
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
100.70s
p50
86.48s
p90
150.29s
p99
204.49s
min
56.95s
max
204.91s
std
41.75s
LLM Time
Total time inside LLM spans, summed across every LLM call in the case.
mean
71.59s
p50
55.56s
p90
115.90s
p99
184.15s
min
33.04s
max
190.23s
std
42.41s
Tool Time
Total time executing tools — database queries, API calls and the like.
mean
15.15s
p50
16.68s
p90
21.93s
p99
26.55s
min
3.72s
max
27.32s
std
6.85s
Other Time
Agent time outside the LLM and tools: orchestration overhead.
mean
2.95s
p50
1.02s
p90
8.94s
p99
12.44s
min
0.00s
max
12.45s
std
4.13s
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
77.8%
p50
78.5%
p90
89.6%
p99
96.2%
min
60.2%
max
96.7%
std
10.4%
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
19.1%
p50
17.3%
p90
31.8%
p99
37.9%
min
4.2%
max
38.6%
std
9.9%
Tool Calls
Number of tool spans the agent produced for the case.
mean
9.72
p50
10.00
p90
11.30
p99
14.49
min
6.00
max
15.00
std
1.99
sum
175.00
Total Tokens
LLM tokens consumed across every call in the case (prompt plus completion).