5
144 forecasters
5 comments
144 forecasters
Will any AI model achieve a score of 95% or higher on the GPQA Diamond Benchmark Leaderboard before June 1, 2026?
ResolvedNo
GPT-5.5 Pro scores below predecessor
Decreases Likelihood
Errors in GPQA benchmark suggest difficulty gap
Decreases Likelihood
GPT-5.5 Pro release may reach 95% threshold
Increases Likelihood