BenchAtlas

Rankings / OpenAI

GPT 5.2 Pro (2025-12-11)

released 2025-12-11

BenchAtlas Index

as of 2026-07-14
Not enough benchmark coverage for a composite score (needs ≥4 families across ≥3 categories). The per-benchmark evidence below stands on its own — sparse coverage is an evidence gap, not a low score.

Benchmark evidence

24 results
external indices
ECImedium effort154.6 pointsindependentEpoch AI Benchmarking Hub
ECIhigh effort154.6 pointsindependentEpoch AI Benchmarking Hub
ECIxhigh effort154.6 pointsindependentEpoch AI Benchmarking Hub
ECI154.6 pointsindependentEpoch AI Benchmarking Hub
reasoning math
ARC-AGI-1older version
split=public_eval · model_type=CoT
xhigh effort97.6 %independentARC Prize Leaderboard
ARC-AGI-1older version
split=public_eval · model_type=CoT
high effort94.6 %independentARC Prize Leaderboard
ARC-AGI-1older version
split=semi_private · model_type=CoT
xhigh effort90.5 %independentARC Prize Leaderboard
ARC-AGI-1older version
split=public_eval · model_type=CoT
medium effort90.3 %independentARC Prize Leaderboard
ARC-AGI-1older version
split=semi_private · model_type=CoT
high effort85.7 %independentARC Prize Leaderboard
ARC-AGI-1older version
split=semi_private · model_type=CoT
medium effort81.2 %independentARC Prize Leaderboard
ARC-AGI-2
split=semi_private · model_type=CoT
high effort54.2 %independentARC Prize Leaderboard
ARC-AGI-2
split=public_eval · model_type=CoT
high effort51.7 %independentARC Prize Leaderboard
ARC-AGI-2
split=semi_private · model_type=CoT
medium effort38.5 %independentARC Prize Leaderboard
ARC-AGI-2
split=public_eval · model_type=CoT
medium effort37.9 %independentARC Prize Leaderboard
ARC-AGI-2
split=public_eval · model_type=CoT
high effort16.7 usd_per_taskindependentARC Prize Leaderboard
ARC-AGI-2
split=semi_private · model_type=CoT
high effort15.7 usd_per_taskindependentARC Prize Leaderboard
ARC-AGI-1older version
split=semi_private · model_type=CoT
xhigh effort11.7 usd_per_taskindependentARC Prize Leaderboard
ARC-AGI-2
split=public_eval · model_type=CoT
medium effort9.5 usd_per_taskindependentARC Prize Leaderboard
ARC-AGI-2
split=semi_private · model_type=CoT
medium effort9.0 usd_per_taskindependentARC Prize Leaderboard
ARC-AGI-1older version
split=public_eval · model_type=CoT
xhigh effort7.7 usd_per_taskindependentARC Prize Leaderboard
ARC-AGI-1older version
split=semi_private · model_type=CoT
high effort5.9 usd_per_taskindependentARC Prize Leaderboard
ARC-AGI-1older version
split=public_eval · model_type=CoT
high effort4.6 usd_per_taskindependentARC Prize Leaderboard
ARC-AGI-1older version
split=semi_private · model_type=CoT
medium effort4.0 usd_per_taskindependentARC Prize Leaderboard
ARC-AGI-1older version
split=public_eval · model_type=CoT
medium effort3.2 usd_per_taskindependentARC Prize Leaderboard