Updated daily at 05:00 UTC
Benchmark
GDPval
Measures: A model produces the deliverables for real tasks written by industry professionals across 44 occupations in 9 U.S. GDP sectors; blinded expert graders compare each output with the professional's own, and the score is the share judged better than (win rate) or as good as (wins or ties) the expert's.
- Maintainer
- OpenAI
- Introduced
- 2025
- Status
- Superseded
- Saturation
- Headroom
- Typical provenance
- Mixed
- In scoring set
- No
Leaderboard
Independent and third-party results only.
No independent or third-party results yet. Show all results