List your company
ResearchFormerly WizWand

Mathematical Reasoning on Math-500 (Throughput, Latency, Score)

13.86Throughputd2Cache
2.64885.55948.4711.3806Sep 27, 2025
Updated 8mo ago

Evaluation results

MethodLinks
d2CacheBackbone=Dream-Base, Few-shot=4-shot, Generation Length=2562025.0913.8618.6339.6
d2CacheBackbone=Dream-Inst, n-shot=4-shot, Generation Length=2562025.0913.818.844.6
d2CacheBackbone=LLaDA-Inst, n-shot=4-shot, Generation Length=2562025.0912.0220.1938
Fast-dLLMBackbone=LLaDA-Base, Few-shot=4-shot, Generation Length=2562025.0910.9323.4127
d2CacheBackbone=LLaDA-Base, Few-shot=4-shot, Generation Length=2562025.0910.820.1330.4
Fast-dLLMBackbone=Dream-Inst, n-shot=4-shot, Generation Length=2562025.0910.7223.5244.4
Fast-dLLMBackbone=Dream-Base, Few-shot=4-shot, Generation Length=2562025.0910.6923.9439.4
Fast-dLLMBackbone=LLaDA-Inst, n-shot=4-shot, Generation Length=2562025.0910.6123.7938
dLLM-CacheBackbone=LLaDA-Base, Few-shot=4-shot, Generation Length=2562025.099.8325.9429.6
dLLM-CacheBackbone=Dream-Base, Few-shot=4-shot, Generation Length=2562025.099.726.0835.2
dLLM-CacheBackbone=Dream-Inst, n-shot=4-shot, Generation Length=2562025.097.1935.3644.2
dLLM-CacheBackbone=LLaDA-Inst, n-shot=4-shot, Generation Length=2562025.096.7137.8438.2
VanillaBackbone=Dream-Base, Few-shot=4-shot, Generation Length=2562025.093.5571.5439
VanillaBackbone=Dream-Inst, n-shot=4-shot, Generation Length=2562025.093.5171.0545.2
VanillaBackbone=LLaDA-Base, Few-shot=4-shot, Generation Length=2562025.093.1480.4432.2
VanillaBackbone=LLaDA-Inst, n-shot=4-shot, Generation Length=2562025.093.0882.5138.4