DeepSeek V3.1 (Reasoning)

deepseek-v3-1-reasoning · 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

26.4 24.4 #1 · 25.4 · 2026-05-11 23:17:33#2 · 25.4 · 2026-05-12 00:00:09#3 · 25.4 · 2026-05-12 01:00:09#4 · 25.4 · 2026-05-12 02:00:11

Cost over time

$485 $483 #1 · $484 · 2026-05-11 23:17:33#2 · $484 · 2026-05-12 00:00:09#3 · $484 · 2026-05-12 01:00:09#4 · $484 · 2026-05-12 02:00:11
Run?Fetched?Score?Quality?Cost?Intel?Code?Agent?MMMU%?
#4 2026-05-12 02:00:11 25.4 25.4 $484 27.7 29.7 18.9 -
#3 2026-05-12 01:00:09 25.4 25.4 $484 27.7 29.7 18.9 -
#2 2026-05-12 00:00:09 25.4 25.4 $484 27.7 29.7 18.9 -
#1 2026-05-11 23:17:33 25.4 25.4 $484 27.7 29.7 18.9 -