List your company
ResearchFormerly WizWand

Short-form Question Answering on SimpleQA (Accuracy)

21.7AccuracyCO-LMLM-360M-FW
-0.7645.06810.916.732Jul 8, 2026
Updated 3mo ago

Evaluation results

MethodLinks
CO-LMLM-360M-FWModel Size=360M, Training Corpus=FW2026.0721.7
CO-LMLM-360MModel Size=360M2026.0721.2
CO-LMLM-135MModel Size=135M2026.0718.2
REL-LMLM-135MModel Size=135M2026.076.2
REL-LMLM-360MModel Size=360M2026.075.6
HF/SMOLLM2-1.7B*Model Size=1.7B, Training Token Count=11T2026.072.1
HF/SMOLLM2-360M*Model Size=360M, Training Token Count=4T2026.071.4
HF/SMOLLM2-135M*Model Size=135M, Training Token Count=2T2026.071.3
STANDARD-360M-FWModel Size=360M, Training Corpus=FW2026.071.3
STANDARD-360MModel Size=360M2026.070.3
STANDARD-135MModel Size=135M2026.070.1