GPT-5.4 (xhigh)

gpt-5-4 · 4 samples · combined / raw

Scoring

Choose the quality benchmark and final score formula for this model.

Cost adjustment

Tune formulas that account for benchmark cost.

Score over time

61.7 59.7 #1 · 60.7 · 2026-05-11 23:17:33#2 · 60.7 · 2026-05-12 00:00:09#3 · 60.7 · 2026-05-12 01:00:09#4 · 60.7 · 2026-05-12 02:00:11

Cost over time

$2,852 $2,850 #1 · $2,851 · 2026-05-11 23:17:33#2 · $2,851 · 2026-05-12 00:00:09#3 · $2,851 · 2026-05-12 01:00:09#4 · $2,851 · 2026-05-12 02:00:11
Run?Fetched?Score?Quality?Cost?Intel?Code?Agent?MMMU%?
#4 2026-05-12 02:00:11 60.7 60.7 $2,851 56.8 57.3 68.0 78.4
#3 2026-05-12 01:00:09 60.7 60.7 $2,851 56.8 57.3 68.0 78.4
#2 2026-05-12 00:00:09 60.7 60.7 $2,851 56.8 57.3 68.0 78.4
#1 2026-05-11 23:17:33 60.7 60.7 $2,851 56.8 57.3 68.0 78.4