List your company
ResearchFormerly WizWand

Reasoning on AIME 25 (Accuracy)

24.17AIME 25 AccuracyDAPO
6.833211.334115.83520.3359Oct 1, 2025
Updated 8mo ago

Evaluation results

MethodLinks
DAPOAlgorithm=DAPO, Stage=Fully Trained, Base Model=Qwen3-8B2025.1024.17
AlphaRL-enhanced DAPOAlgorithm=DAPO, Stage=40%, Base Model=Qwen3-8B2025.1023.75
GRPOAlgorithm=GRPO, Stage=Fully Trained, Base Model=Qwen3-8B2025.1021.67
RLOOAlgorithm=RLOO, Stage=Fully Trained, Base Model=Qwen3-8B2025.1018.33
AlphaRL-enhanced RLOOAlgorithm=RLOO, Stage=40%, Base Model=Qwen3-8B2025.1018.33
AlphaRL-enhanced GRPOAlgorithm=GRPO, Stage=40%, Base Model=Qwen3-8B2025.1018.13
GRPOAlgorithm=GRPO, Stage=40%, Base Model=Qwen3-8B2025.1014.17
RLOOAlgorithm=RLOO, Stage=40%, Base Model=Qwen3-8B2025.1014.17
AlphaRL-enhanced RLOOAlgorithm=RLOO, Stage=10%, Base Model=Qwen3-8B2025.1014.17
AlphaRL-enhanced GRPOAlgorithm=GRPO, Stage=10%, Base Model=Qwen3-8B2025.1013.25
DAPOAlgorithm=DAPO, Stage=40%, Base Model=Qwen3-8B2025.1011.67
AlphaRL-enhanced DAPOAlgorithm=DAPO, Stage=10%, Base Model=Qwen3-8B2025.1011.67
GRPOAlgorithm=GRPO, Stage=10%, Base Model=Qwen3-8B2025.108.33
RLOOAlgorithm=RLOO, Stage=10%, Base Model=Qwen3-8B2025.108.33
DAPOAlgorithm=DAPO, Stage=10%, Base Model=Qwen3-8B2025.107.5