List your company

Reasoning on Sudoku

96Pass@1BoN w. ReJump
1.8826.31550.7575.185Nov 30, 2025Dec 30, 2025Jan 30, 2026Mar 2, 2026Apr 2, 2026May 3, 2026Jun 3, 2026

Evaluation results

MethodLinks
BoN w. ReJumpModel=Grok 3 Mini Beta2025.1196——0.71—
SPG w/ MixtureSeq Len=2562026.0394————
SPG w/ MixtureSeq Len=5122026.0393.1————
CAPRBackbone=LLaDA-8B-Instruct, Decoding Budget (L)=256, Zero-shot setting=true2026.0693————
d-TreeRPOBackbone=LLaDA-8B-Instruct, Decoding Budget (L)=256, Zero-shot setting=true2026.0692.9————
ESPOSeq Len=1282026.0392.7————
CAPRBackbone=LLaDA-8B-Instruct, Decoding Budget (L)=512, Zero-shot setting=true2026.0692.7————
StableDRLSeq Len=2562026.0392.4————
StableDRLSeq Len=1282026.0391.9————
Majority VoteModel=Grok 3 Mini Beta2025.1191——6.01—
StableDRLSeq Len=5122026.0390.1————
CAPRBackbone=LLaDA-MoE-7B-A1B-Instruct, Decoding Budget (L)=256, Zero-shot setting=true2026.0684.8————
ESPOSeq Len=2562026.0384.7————
SPG w/ MixtureSeq Len=1282026.0382.9————
CAPRBackbone=LLaDA-MoE-7B-A1B-Instruct, Decoding Budget (L)=512, Zero-shot setting=true2026.0681————
ESPOSeq Len=5122026.0380.5————
d-TreeRPOBackbone=LLaDA-8B-Instruct, Decoding Budget (L)=512, Zero-shot setting=true2026.0680.3————
d-TreeRPOBackbone=LLaDA-MoE-7B-A1B-Instruct, Decoding Budget (L)=256, Zero-shot setting=true2026.0678.3————
d2-stepMergeBackbone=LLaDA-8B-Instruct, Decoding Budget (L)=256, Zero-shot setting=true2026.0676.1————
d-TreeRPOBackbone=LLaDA-MoE-7B-A1B-Instruct, Decoding Budget (L)=512, Zero-shot setting=true2026.0672————
UniGRPOSeq Len=2562026.0367————
d2-stepMergeBackbone=LLaDA-8B-Instruct, Decoding Budget (L)=512, Zero-shot setting=true2026.0666.2————
UniGRPOSeq Len=5122026.0362.9————
UniGRPOSeq Len=1282026.0359————
wd1Backbone=LLaDA-MoE-7B-A1B-Instruct, Decoding Budget (L)=256, Zero-shot setting=true2026.0640.6————
wd1Backbone=LLaDA-MoE-7B-A1B-Instruct, Decoding Budget (L)=512, Zero-shot setting=true2026.0635.4————
WD1Seq Len=1282026.0333.1————
D1Seq Len=2562026.0332.5————
WD1Seq Len=2562026.0332.1————
GDPOBackbone=LLaDA-MoE-7B-A1B-Instruct, Decoding Budget (L)=256, Zero-shot setting=true2026.0631.2————
GDPOBackbone=LLaDA-MoE-7B-A1B-Instruct, Decoding Budget (L)=512, Zero-shot setting=true2026.0630.6————
D1Seq Len=5122026.0329.3————
LLaDA-1.5Seq Len=5122026.0329————
LLaDA-8B-Inst.Seq Len=2562026.0327.7————
TraceRLBackbone=LLaDA-MoE-7B-A1B-Instruct, Decoding Budget (L)=256, Zero-shot setting=true2026.0627.4————
LLaDA-1.5Seq Len=2562026.0326.9————
LLaDA-8B-Inst.Seq Len=5122026.0326.2————
TraceRLBackbone=LLaDA-8B-Instruct, Decoding Budget (L)=256, Zero-shot setting=true2026.0625.6————
TraceRLBackbone=LLaDA-8B-Instruct, Decoding Budget (L)=512, Zero-shot setting=true2026.0625.4————
wd1Backbone=LLaDA-8B-Instruct, Decoding Budget (L)=256, Zero-shot setting=true2026.0625.2————
TraceRLBackbone=LLaDA-MoE-7B-A1B-Instruct, Decoding Budget (L)=512, Zero-shot setting=true2026.0625.2————
wd1Backbone=LLaDA-8B-Instruct, Decoding Budget (L)=512, Zero-shot setting=true2026.0624.2————
WD1Seq Len=5122026.0322.5————
SAPOBackbone=LLaDA-MoE-7B-A1B-Instruct, Decoding Budget (L)=256, Zero-shot setting=true2026.0621.2————
SAPOBackbone=LLaDA-8B-Instruct, Decoding Budget (L)=256, Zero-shot setting=true2026.0620.3————
SAPOBackbone=LLaDA-MoE-7B-A1B-Instruct, Decoding Budget (L)=512, Zero-shot setting=true2026.0618.7————
Diffu-GRPOBackbone=LLaDA-MoE-7B-A1B-Instruct, Decoding Budget (L)=256, Zero-shot setting=true2026.0617.3————
SAPOBackbone=LLaDA-8B-Instruct, Decoding Budget (L)=512, Zero-shot setting=true2026.0616.1————
Diffu-GRPOBackbone=LLaDA-MoE-7B-A1B-Instruct, Decoding Budget (L)=512, Zero-shot setting=true2026.0615.4————
Diffu-GRPOBackbone=LLaDA-8B-Instruct, Decoding Budget (L)=256, Zero-shot setting=true2026.0612.9————
VRPOBackbone=LLaDA-8B-Instruct, Decoding Budget (L)=256, Zero-shot setting=true2026.0612.8————
BaseBackbone=LLaDA-MoE-7B-A1B-Instruct, Decoding Budget (L)=256, Zero-shot setting=true2026.0612.7————
Diffu-GRPOBackbone=LLaDA-8B-Instruct, Decoding Budget (L)=512, Zero-shot setting=true2026.0611.2————
VRPOBackbone=LLaDA-8B-Instruct, Decoding Budget (L)=512, Zero-shot setting=true2026.069.6————
BaseBackbone=LLaDA-MoE-7B-A1B-Instruct, Decoding Budget (L)=512, Zero-shot setting=true2026.068.1————
LLaDA-1.5Seq Len=1282026.037.4————
D1Seq Len=1282026.037.2————
BaseBackbone=LLaDA-8B-Instruct, Decoding Budget (L)=256, Zero-shot setting=true2026.066.7————
LLaDA-8B-Inst.Seq Len=1282026.035.7————
BaseBackbone=LLaDA-8B-Instruct, Decoding Budget (L)=512, Zero-shot setting=true2026.065.5————
Baseline (Full Steps)Sequence Length=1282025.11—64———
Baseline (Full Steps)Sequence Length=2562025.11—128———
Baseline (Full Steps)Sequence Length=5122025.11—256———
D1Seq Len=Average2026.03————23
EDITSequence Length=1282025.11—38.340.2——
EDITSequence Length=2562025.11—74.941.5——
EDITSequence Length=5122025.11—163.336.2——
ESPOSeq Len=Average2026.03————86
LLaDA-1.5Seq Len=Average2026.03————21.1
LLaDA-8B-Inst.Seq Len=Average2026.03————19.9
SPG w/ MixtureSeq Len=Average2026.03————90
StableDRLSeq Len=Average2026.03————91.5
UniGRPOSeq Len=Average2026.03————63
WD1Seq Len=Average2026.03————29.2