Run started 2026-10-18 07:59 UTC
· completed 08:05 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
75.21s
p50
74.98s
p90
84.94s
p99
87.62s
min
62.95s
max
87.92s
std
10.49s
First Insight Time
Time to the agent's first update, such as a tool call or a reasoning step.
mean
10.79s
p50
10.35s
p90
12.61s
p99
13.24s
min
9.17s
max
13.31s
std
1.84s
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
74.05s
p50
82.16s
p90
103.12s
p99
122.49s
min
11.40s
max
124.64s
std
36.20s
LLM Time
Total time inside LLM spans, summed across every LLM call in the case.
mean
50.06s
p50
48.01s
p90
62.88s
p99
68.22s
min
35.43s
max
68.81s
std
13.86s
Tool Time
Total time executing tools — database queries, API calls and the like.
mean
31.19s
p50
29.02s
p90
40.48s
p99
44.67s
min
21.58s
max
45.14s
std
9.95s
Other Time
Agent time outside the LLM and tools: orchestration overhead.
mean
0.87s
p50
0.00s
p90
2.44s
p99
3.39s
min
0.00s
max
3.49s
std
1.75s
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
63.5%
p50
66.2%
p90
75.8%
p99
76.6%
min
45.0%
max
76.7%
std
14.7%
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
40.3%
p50
39.9%
p90
52.9%
p99
56.9%
min
24.1%
max
57.3%
std
13.8%
Tool Calls
Number of tool spans the agent produced for the case.
mean
13.00
p50
13.00
p90
13.70
p99
13.97
min
12.00
max
14.00
std
0.82
sum
52.00
Total Tokens
LLM tokens consumed across every call in the case (prompt plus completion).