OpenAI
GPT-5.2
Released 2025-12-11 · Context 400K · Data snapshot 2026-10-26
Overall score
58.9
Ranked #17Benchmark coverage
26.9%
7 third-party benchmarks
Context window
400K
tokens
Input price
$1.75
Cache read $0.18
Output price
$14.00
per million tokens
OpenRouter usage
#107
54.9B tokens/day
Prices converted at snapshot-day rate (USD/CNY 6.6702)
Dimension performance
Human preference
62.7
Office
60.6
Coding
57.5
Above averageBelow averageAnchored at 50 (reference models’ average); every 15 points ≈ one standard deviation
Benchmark details
Coding2 items
Terminal-Bench64.9
Vibe Code Bench53.5
Office3 items
Tax Eval v275.76
Corporate Finance v265.89
MedCode49.75
Knowledge1 items
MMLU-Pro86.23
Human preference1 items
LMArena Human Blind Vote1435.4701
Multiple reasoning tiers of the same benchmark take the best value; raw scores have different scales and are not comparable across benchmarks.