Benchmark · KnowledgeCompetitive

Gdpval

Updated 2025-12-10
Models tested
11
Top score
49.7
GPT-5.2
Median
34.8
min 9.9
Top-5 spread
σ 3.2
Competitive

Best score over time · one chart, every benchmark

GDPVAL9 MODELS · FRONTIER RUNNING MAX0255075100SCORE ↑Nov 24Feb 25May 25Sep 25Dec 25RELEASE DATE →benchgecko.ai/benchmark/gdpval · frontier
Frontier on Gdpval rose from 9.9 to 49.7 in 13 months · +39.8 points · latest leader GPT-5.2 from OpenAI.
Pink dots = frontier records · 5 totalClick to open model page

11 models tested · sorted by score

Same category · related evaluations