Professional accounting / Benchmark leaderboard
APEX-Accounting
Month-end accounting work across 10 business worlds: 160 tasks and 2,186 grading criteria.
No published company result yet.
This benchmark is in the research directory. Its task package, adapter and grading protocol need qualification before a hosted run can be offered.
Arrange an evaluation for your company →What this benchmark measures
Metrics
Mean criteria passed; task success and consistency across repeated attempts. Publisher reports task-level bootstrap intervals.
Execution requirements
Accounting applications, spreadsheets, documents and code; source-defined task rubrics and the published Loop Harness.
Scope and limitations
The 160-task leaderboard split is private. A separate public world is a sample, not the full benchmark. Licensed source access is required before full-suite execution.
Evaluation availability
Catalog entry. Request a managed evaluation to qualify your agent interface and the benchmark’s native grading requirements.
Sources and company fit
- Original benchmark ↗Publisher benchmark · full split requires access
- Publisher · Mercor APEX-Accounting ↗160 source tasks · Private evaluation split; separate public sample · Publisher methodology checked 2026-09-16
- Publisher · Public sample task and grading rubric ↗One public sample world; excluded from the private evaluation split · Publisher sample inspected 2026-09-16
- Publisher · Public sample trajectory and response ↗One public sample world; excluded from the private evaluation split · Publisher sample inspected 2026-09-16