Claude 4.5 Haiku (Reasoning)

claude-4-5-haiku-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

37.6 35.6 #1 · 36.6 · 2026-05-11 23:17:33#2 · 36.6 · 2026-05-12 00:00:09#3 · 36.6 · 2026-05-12 01:00:09#4 · 36.6 · 2026-05-12 02:00:11

Cost over time

$621 $619 #1 · $620 · 2026-05-11 23:17:33#2 · $620 · 2026-05-12 00:00:09#3 · $620 · 2026-05-12 01:00:09#4 · $620 · 2026-05-12 02:00:11
Run?Fetched?Score?Quality?Cost?Intel?Code?Agent?MMMU%?
#4 2026-05-12 02:00:11 36.6 36.6 $620 37.1 32.6 40.2 58.6
#3 2026-05-12 01:00:09 36.6 36.6 $620 37.1 32.6 40.2 58.6
#2 2026-05-12 00:00:09 36.6 36.6 $620 37.1 32.6 40.2 58.6
#1 2026-05-11 23:17:33 36.6 36.6 $620 37.1 32.6 40.2 58.6