Run started 2026-10-19 23:31 UTC
· completed 00:11 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
77.15s
p50
63.56s
p90
144.18s
p99
183.54s
min
8.41s
max
194.56s
std
40.39s
First Insight Time
Time to the agent's first update, such as a tool call or a reasoning step.
mean
9.88s
p50
7.72s
p90
14.90s
p99
41.59s
min
2.41s
max
56.13s
std
8.08s
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
93.84s
p50
81.03s
p90
161.76s
p99
204.11s
min
17.13s
max
207.45s
std
43.47s
LLM Time
Total time inside LLM spans, summed across every LLM call in the case.
mean
58.51s
p50
51.25s
p90
95.93s
p99
134.87s
min
7.39s
max
138.69s
std
28.60s
Tool Time
Total time executing tools — database queries, API calls and the like.
mean
20.20s
p50
15.39s
p90
37.58s
p99
80.06s
min
0.21s
max
81.92s
std
17.94s
Other Time
Agent time outside the LLM and tools: orchestration overhead.
mean
3.56s
p50
0.70s
p90
9.09s
p99
31.39s
min
0.00s
max
47.54s
std
7.36s
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
74.5%
p50
78.0%
p90
86.9%
p99
92.2%
min
36.4%
max
93.4%
std
12.3%
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
22.9%
p50
21.2%
p90
36.4%
p99
56.1%
min
2.5%
max
56.6%
std
11.3%
Tool Calls
Number of tool spans the agent produced for the case.
mean
9.73
p50
9.00
p90
13.20
p99
18.56
min
1.00
max
20.00
std
3.34
sum
477.00
Total Tokens
LLM tokens consumed across every call in the case (prompt plus completion).