Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Grade-school mathematical reasoning on GSM8K (Throughput)

160.51Output Throughput

Our approach

155.2684156.6292157.99159.3508Mar 5, 2026
Updated 2mo ago

Evaluation Results

MethodLinks
2026.03
160.513.2
2026.03
155.47-