List your company

Mathematical Reasoning on AIME24 (Accuracy, Generation Length)

29.2AccuracyDeepSeek-R1-1.5B
-1.1686.71614.622.484Jan 7, 2026

Evaluation results

MethodLinks
DeepSeek-R1-1.5BReasoning Strategy=Long-CoT2026.0129.217,465
RPAMMethod Category=Data-dependent Merging2026.0126.74,984
CoDMethod Category=Prompt-guided2026.01208,426
Ada-R1Method Category=Training-based2026.0116.73,931
Task ArithmeticMethod Category=Data-free Merging2026.0116.73,158
AIMMethod Category=Data-dependent Merging2026.0116.74,022
Average MergingMethod Category=Data-free Merging2026.0113.33,191
TIES MergingMethod Category=Data-free Merging2026.01102,901
DARE-LinearMethod Category=Data-free Merging2026.01102,451
ACMMethod Category=Data-dependent Merging2026.01102,689
Qwen2.5-Math-1.5BReasoning Strategy=Short-CoT2026.010865