List your company
ResearchFormerly WizWand

General Reasoning on SuperGPQA (Accuracy)

37.77Accuracy (General Reasoning)INFUSER
21.733225.896630.0634.2234May 16, 2026May 19, 2026May 23, 2026May 27, 2026May 31, 2026Jun 4, 2026Jun 8, 2026
Updated 4mo ago

Evaluation results

MethodLinks
INFUSERModel Backbone=Qwen3-8B-Base2026.0637.77
+ SPICEBackbone=Qwen3-8B-Base, #Data=20K2026.0535.7
SPICEModel Backbone=Qwen3-8B-Base2026.0635.7
D2EvoBackbone=Qwen3-8B-Base, #Data=1.7K, Iteration=32026.0533.71
+ AZRBackbone=Qwen3-8B-Base2026.0533.5
R-FewModel Backbone=Qwen3-8B-Base2026.0633.5
INFUSERModel Backbone=Qwen3-4B-Base2026.0633.48
+ Full DataBackbone=Qwen3-8B-Base, #Data=19K2026.0532.84
AZRModel Backbone=Qwen3-8B-Base2026.0632.63
+ R-ZeroBackbone=Qwen3-8B-Base2026.0532.1
R-ZeroModel Backbone=Qwen3-8B-Base2026.0632.06
Base ModelBackbone=Qwen3-8B-Base2026.0531.06
BaseModel Backbone=Qwen3-8B-Base2026.0630.62
+ SPICEBackbone=Qwen3-4B-Base, #Data=20K2026.0530.2
SPICEModel Backbone=Qwen3-4B-Base2026.0630.2
+ Full DataBackbone=Qwen3-4B-Base, #Data=19K2026.0529.6
D2EvoBackbone=Qwen3-4B-Base, #Data=1.4K, Iteration=32026.0529.51
R-FewModel Backbone=Qwen3-4B-Base2026.0629.4
R-ZeroModel Backbone=Qwen3-4B-Base2026.0628.4
AZRModel Backbone=Qwen3-4B-Base2026.0628.31
+ R-ZeroBackbone=Qwen3-4B-Base2026.0527.96
+ AZRBackbone=Qwen3-4B-Base2026.0527.1
BaseModel Backbone=Qwen3-4B-Base2026.0625.88
Base ModelBackbone=Qwen3-4B-Base2026.0525.23
D2EvoBackbone=Llama-3.1-8B, #Data=1.2K, Iteration=32026.0525.18
+ Full DataBackbone=Llama-3.1-8B, #Data=19K2026.0524.53
Base ModelBackbone=Llama-3.1-8B2026.0522.35