List your company

Mathematical Reasoning on Average (GSM8K, MATH500, AIME24)

6.73Latency (s)STACK
6.35368.894311.43513.9757Apr 10, 2026

Evaluation results

MethodLinks
STACKBackbone=DeepSeek-R1-Distill-Qwen-7B2026.046.732.7-54.5
STACKBackbone=DeepSeek-R1-Distill-Qwen-1.5B2026.047.234.8-59.9
MuTISBackbone=DeepSeek-R1-Distill-Qwen-7B2026.047.951.03-53.9
ConCISEBackbone=DeepSeek-R1-Distill-Qwen-7B2026.048.6-2.23-46.8
MuTISBackbone=DeepSeek-R1-Distill-Qwen-1.5B2026.048.832.07-56.7
ConCISEBackbone=DeepSeek-R1-Distill-Qwen-1.5B2026.049.550.83-51.5
PromptBackbone=DeepSeek-R1-Distill-Qwen-7B2026.0412.470.6-15.1
OriginalBackbone=DeepSeek-R1-Distill-Qwen-7B2026.0413.29——
PromptBackbone=DeepSeek-R1-Distill-Qwen-1.5B2026.0415.10.77-7.35
OriginalBackbone=DeepSeek-R1-Distill-Qwen-1.5B2026.0416.14——