Run started 2026-10-18 08:09 UTC
· completed 08:14 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
96.45s
p50
94.86s
p90
118.85s
p99
126.43s
min
68.80s
max
127.27s
std
24.20s
First Insight Time
Time to the agent's first update, such as a tool call or a reasoning step.
mean
10.96s
p50
12.26s
p90
12.36s
p99
12.39s
min
6.94s
max
12.39s
std
2.68s
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
67.54s
p50
74.01s
p90
106.00s
p99
130.66s
min
7.60s
max
133.41s
std
39.99s
LLM Time
Total time inside LLM spans, summed across every LLM call in the case.
mean
70.34s
p50
73.12s
p90
80.17s
p99
80.92s
min
54.09s
max
81.00s
std
12.19s
Tool Time
Total time executing tools — database queries, API calls and the like.
mean
27.98s
p50
22.74s
p90
44.52s
p99
52.90s
min
12.63s
max
53.83s
std
17.88s
Other Time
Agent time outside the LLM and tools: orchestration overhead.
mean
2.00s
p50
0.92s
p90
4.87s
p99
6.04s
min
0.00s
max
6.17s
std
2.91s
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
72.7%
p50
74.4%
p90
79.4%
p99
81.0%
min
60.7%
max
81.2%
std
8.6%
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
27.7%
p50
28.1%
p90
38.7%
p99
41.5%
min
12.7%
max
41.8%
std
12.2%
Tool Calls
Number of tool spans the agent produced for the case.
mean
12.00
p50
13.00
p90
13.00
p99
13.00
min
9.00
max
13.00
std
2.00
sum
48.00
Total Tokens
LLM tokens consumed across every call in the case (prompt plus completion).