Tax Agent Bench

Measures: An agent researches research-grade US corporate tax questions (about 60% federal corporate income tax, plus partnership, SALT, transfer pricing) with tools, then submits one cited answer; scored as a weighted rubric score (zero if any must-pass check fails) multiplied by a citation-quality factor.

Maintainer: Vals AI (questions and rubrics by tax professionals)
Introduced: 2026
Status: Active
Saturation: Headroom
Typical provenance: Independent
In scoring set: Yes

Official leaderboard

Leaderboard

Show independent results only

Every result we hold, whoever ran it. Only independent and third-party rows count towards a score.

No results to show.