Run started 2026-10-18 08:27 UTC
· completed 08:33 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
92.93s
p50
81.72s
p90
145.83s
p99
170.95s
min
40.96s
max
173.74s
std
38.52s
First Insight Time
Time to the agent's first update, such as a tool call or a reasoning step.
mean
10.30s
p50
10.83s
p90
13.30s
p99
13.76s
min
4.22s
max
13.81s
std
2.84s
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
102.44s
p50
94.83s
p90
153.69s
p99
183.62s
min
48.81s
max
186.95s
std
38.85s
LLM Time
Total time inside LLM spans, summed across every LLM call in the case.
mean
54.40s
p50
46.73s
p90
95.69s
p99
98.08s
min
25.62s
max
98.34s
std
25.30s
Tool Time
Total time executing tools — database queries, API calls and the like.
mean
45.78s
p50
44.75s
p90
74.67s
p99
85.06s
min
11.43s
max
86.21s
std
21.32s
Other Time
Agent time outside the LLM and tools: orchestration overhead.
mean
1.14s
p50
0.00s
p90
2.40s
p99
6.62s
min
0.00s
max
7.09s
std
2.22s
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
56.7%
p50
58.2%
p90
68.4%
p99
75.2%
min
37.1%
max
76.0%
std
11.0%
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
49.9%
p50
48.0%
p90
67.8%
p99
85.4%
min
14.8%
max
87.3%
std
19.6%
Tool Calls
Number of tool spans the agent produced for the case.
mean
11.70
p50
11.00
p90
16.30
p99
18.73
min
7.00
max
19.00
std
3.89
sum
117.00
Total Tokens
LLM tokens consumed across every call in the case (prompt plus completion).