DeepSeek V3.2 Exp (Reasoning)

deepseek-v3-2-reasoning-0925 · 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

32.7 30.7 #1 · 31.7 · 2026-05-11 23:17:33#2 · 31.7 · 2026-05-12 00:00:09#3 · 31.7 · 2026-05-12 01:00:09#4 · 31.7 · 2026-05-12 02:00:11

Cost over time

$40.1 $38.1 #1 · $39.1 · 2026-05-11 23:17:33#2 · $39.1 · 2026-05-12 00:00:09#3 · $39.1 · 2026-05-12 01:00:09#4 · $39.1 · 2026-05-12 02:00:11
Run?Fetched?Score?Quality?Cost?Intel?Code?Agent?MMMU%?
#4 2026-05-12 02:00:11 31.7 31.7 $39.1 32.9 33.3 28.7 -
#3 2026-05-12 01:00:09 31.7 31.7 $39.1 32.9 33.3 28.7 -
#2 2026-05-12 00:00:09 31.7 31.7 $39.1 32.9 33.3 28.7 -
#1 2026-05-11 23:17:33 31.7 31.7 $39.1 32.9 33.3 28.7 -