List your company
ResearchFormerly WizWand

Mathematics Reasoning on GSM8K (Accuracy)

91.3AccuracySDAR-8B
78.09281.52184.9588.379Jun 2, 2026Jun 3, 2026Jun 4, 2026
Updated 4mo ago

Evaluation results

MethodLinks
SDAR-8BScale=8B, Tokens=55B, FLOPs=2640, Decoding Strategy=greedy static decoding, Block Size=42026.0691.3
SDAR-4BScale=4B, Tokens=55B, FLOPs=1320, Decoding Strategy=greedy static decoding, Block Size=42026.0689.9
Fisher Merging w/ HARCMerging Strategy=Fisher Merging, Model Architecture=Qwen3-30B-A3B, HARC Calibration=true2026.0687.99
IndividualModel Architecture=Qwen3-30B-A3B, HARC Calibration=false2026.0687.97
Fisher MergingMerging Strategy=Fisher Merging, Model Architecture=Qwen3-30B-A3B, HARC Calibration=false2026.0687.95
Weight Averaging w/ HARCMerging Strategy=Weight Averaging, Model Architecture=Qwen3-30B-A3B, HARC Calibration=true2026.0687.91
Weight AveragingMerging Strategy=Weight Averaging, Model Architecture=Qwen3-30B-A3B, HARC Calibration=false2026.0687.86
TIES-Merging w/ HARCMerging Strategy=TIES-Merging, Model Architecture=Qwen3-30B-A3B, HARC Calibration=true2026.0687.76
DAREMerging Strategy=DARE, Model Architecture=Qwen3-30B-A3B, HARC Calibration=false2026.0687.65
TIES-MergingMerging Strategy=TIES-Merging, Model Architecture=Qwen3-30B-A3B, HARC Calibration=false2026.0687.64
DARE w/ HARCMerging Strategy=DARE, Model Architecture=Qwen3-30B-A3B, HARC Calibration=true2026.0687.61
OPDLM-4BScale=4B, Tokens=0.076B, FLOPs=2.4, Decoding Strategy=greedy static decoding, Block Size=42026.0687.6
WUDI-Merging w/ HARCMerging Strategy=WUDI-Merging, Model Architecture=Qwen3-30B-A3B, HARC Calibration=true2026.0687.53
WUDI-MergingMerging Strategy=WUDI-Merging, Model Architecture=Qwen3-30B-A3B, HARC Calibration=false2026.0687.34
OPDLM-8BScale=8B, Tokens=0.066B, FLOPs=4.2, Decoding Strategy=greedy static decoding, Block Size=42026.0687.1
Fast-dLLM-v2-7BScale=8B, Tokens=1B, FLOPs=42, Decoding Strategy=greedy static decoding, Block Size=42026.0683.7
Dream-7BScale=8B, Tokens=580B, FLOPs=24360, Decoding Strategy=greedy static decoding, Block Size=42026.0681
LLaDA-8BScale=8B, Tokens=1500B, FLOPs=72000, Decoding Strategy=greedy static decoding, Block Size=42026.0678.6