Run started 2026-10-18 10:18 UTC
· completed 10:38 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
83.19s
p50
74.17s
p90
114.49s
p99
127.08s
min
51.71s
max
128.66s
std
24.96s
First Insight Time
Time to the agent's first update, such as a tool call or a reasoning step.
mean
8.35s
p50
5.49s
p90
13.18s
p99
32.20s
min
2.52s
max
35.20s
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
100.79s
p50
86.02s
p90
132.79s
p99
188.06s
min
62.95s
max
196.47s
std
35.31s
LLM Time
Total time inside LLM spans, summed across every LLM call in the case.
mean
55.95s
p50
52.98s
p90
77.43s
p99
84.74s
min
37.30s
max
85.77s
std
14.76s
Tool Time
Total time executing tools — database queries, API calls and the like.
mean
31.23s
p50
25.44s
p90
54.20s
p99
76.21s
min
6.35s
max
79.18s
std
19.21s
Other Time
Agent time outside the LLM and tools: orchestration overhead.
mean
0.30s
p50
0.00s
p90
0.50s
p99
3.16s
min
0.00s
max
3.56s
std
0.92s
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
68.1%
p50
68.0%
p90
86.0%
p99
93.9%
min
39.7%
max
94.8%
std
14.9%
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
34.7%
p50
37.7%
p90
51.2%
p99
60.0%
min
11.7%
max
60.8%
std
14.0%
Tool Calls
Number of tool spans the agent produced for the case.
mean
11.53
p50
12.00
p90
14.60
p99
15.86
min
6.00
max
16.00
std
2.70
sum
173.00
Total Tokens
LLM tokens consumed across every call in the case (prompt plus completion).