Class DemoBenchmark

All Implemented Interfaces:
GuardedExecution, ScopeAuthority, Identifiable, Job<DemoAnswer>, Doer<DemoQuery, DemoAnswer>, Orchestrator<DemoQuery, DemoAnswer>, Tool<DemoQuery, DemoAnswer>

public final class DemoBenchmark extends Benchmark<DemoQuery, DemoAnswer>
The demo's benchmark of DemoAgent: the reference answers on the strongest model the deployment serves - resolved through Grade.CEILING, the same way the judge is - while every raced copy is re-pinned to its candidate, and the judges wait behind the same person the raced agents do. A named class, so its job ids and its log lines (ai.redouble.demo.DemoBenchmark.progress for the runs-finished count) name it.
Since:
0.1 (2026-09-22)
Author:
Andrey Santrosyan
  • Constructor Details

    • DemoBenchmark

      public DemoBenchmark(Identifiable workflow, int runs)
      Parameters:
      workflow - the workflow the benchmark and every agent it builds belong to
      runs - how many times each candidate answers
  • Method Details