GDPval-AA

Measures: An agent with shell and web access (Artificial Analysis's Stirrup harness) produces real work deliverables for OpenAI's 220 public GDPval tasks across 44 occupations; each pair of model outputs is compared blind by a panel of three LLM judges and the results are fitted to an Elo scale.

Maintainer: Artificial Analysis (task set from OpenAI's GDPval)
Introduced: —
Status: Active
Saturation: Headroom
Typical provenance: Independent
In scoring set: Yes

Paper 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.