List your company

Reasoning on Sudoku (Accuracy)

91.9Accuracy (Sudoku Reasoning)StepMerge (d2)
4.5427.2249.972.58Sep 25, 2025Nov 3, 2025Dec 13, 2025Jan 22, 2026Mar 2, 2026Apr 11, 2026May 21, 2026

Evaluation results

MethodLinks
StepMerge (d2)Post-training=true2025.0991.9
wd1Post-training=true, Evaluation Source=Corresponding paper2025.0925.2
diffu-GRPO (d1)Post-training=true, Evaluation Source=Corresponding paper2025.0922.1
LIFT3Backbone=LLaDA-1.52026.0518.2
GIFTBackbone=LLaDA-1.52026.0517.6
LIFT3Backbone=LLaDA-8B-Instruct2026.0517.4
GIFTBackbone=LLaDA-8B-Instruct2026.0517.3
DoRA†Backbone=LLaDA-8B-Instruct, LoRA Rank=128, Tuned Params (Usage %)=101.1M (1.25%), Steps=128, Zero-shot=true2026.0317.2
CARTBackbone=LLaDA-1.52026.0517
VanillaBackbone=LLaDA-8B-Instruct2026.0516.8
LoRABackbone=LLaDA-8B-Instruct, LoRA Rank=128, Tuned Params (Usage %)=100.7M (1.24%), Steps=128, Zero-shot=true2026.0316.5
LIFT2Backbone=LLaDA-8B-Instruct2026.0516.5
LIFT2Backbone=LLaDA-1.52026.0515.6
CARTBackbone=LLaDA-8B-Instruct2026.0514.6
LoRA†Backbone=LLaDA-8B-Instruct, LoRA Rank=64, Tuned Params (Usage %)=50.3M (0.62%), Steps=128, Zero-shot=true2026.0314.5
VanillaBackbone=LLaDA-1.52026.0514.4
LLaDA 1.5Post-training=false, Evaluation Source=Released checkpoint2025.0912.5
LLaDA 1.5Backbone=LLaDA-1.52026.0512.1
LLaDAPost-training=false, Evaluation Source=Released checkpoint2025.0911.8
LLaDABackbone=LLaDA-8B-Instruct2026.0511.2
AlfaBackbone=LLaDA-8B-Instruct, LoRA Rank=64, Tuned Params (Usage %)=69.2M (0.85%), Steps=128, Zero-shot=true2026.039.7
LoRABackbone=LLaDA-8B-Instruct, LoRA Rank=128, Tuned Params (Usage %)=100.7M (1.24%), Steps=256, Zero-shot=true2026.038.5
AlfaBackbone=LLaDA-8B-Instruct, LoRA Rank=64, Tuned Params (Usage %)=69.2M (0.85%), Steps=256, Zero-shot=true2026.038.3
DoRA†Backbone=LLaDA-8B-Instruct, LoRA Rank=128, Tuned Params (Usage %)=101.1M (1.25%), Steps=256, Zero-shot=true2026.038
LoRA†Backbone=LLaDA-8B-Instruct, LoRA Rank=64, Tuned Params (Usage %)=50.3M (0.62%), Steps=256, Zero-shot=true2026.037.9