Run started 2026-10-18 09:45 UTC
· completed 09:59 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
86.15s
p50
79.61s
p90
115.35s
p99
137.16s
min
54.23s
max
139.58s
std
27.02s
First Insight Time
Time to the agent's first update, such as a tool call or a reasoning step.
mean
9.76s
p50
7.60s
p90
16.22s
p99
16.82s
min
4.72s
max
16.89s
std
4.51s
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
100.19s
p50
96.36s
p90
133.27s
p99
149.82s
min
61.36s
max
151.81s
std
28.57s
LLM Time
Total time inside LLM spans, summed across every LLM call in the case.
mean
67.01s
p50
61.55s
p90
86.60s
p99
124.21s
min
41.47s
max
128.39s
std
26.21s
Tool Time
Total time executing tools — database queries, API calls and the like.
mean
24.94s
p50
21.74s
p90
39.32s
p99
45.76s
min
6.18s
max
46.47s
std
12.02s
Other Time
Agent time outside the LLM and tools: orchestration overhead.
mean
0.18s
p50
0.00s
p90
0.59s
p99
1.34s
min
0.00s
max
1.42s
std
0.45s
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
75.5%
p50
79.6%
p90
89.3%
p99
95.2%
min
56.7%
max
95.9%
std
13.2%
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
28.7%
p50
22.4%
p90
43.8%
p99
48.5%
min
10.8%
max
49.0%
std
13.1%
Tool Calls
Number of tool spans the agent produced for the case.
mean
12.27
p50
13.00
p90
14.00
p99
14.00
min
8.00
max
14.00
std
1.85
sum
135.00
Total Tokens
LLM tokens consumed across every call in the case (prompt plus completion).