List your company

Mathematical Reasoning on IMO-Bench

57.02AccuracyF-1
0.423215.116629.8144.5034Jan 27, 2026Feb 15, 2026Mar 6, 2026Mar 25, 2026Apr 13, 2026May 2, 2026May 21, 2026

Evaluation results

MethodLinks
F-1Model=Gemini 2.5 Pro, Model Category=Proprietary2026.0157.02
Zero-ShotModel=GPT-5, Model Category=Proprietary2026.0156.26
F-1Model=GPT-5, Model Category=Proprietary2026.0155.58
Zero-ShotModel=Gemini 2.5 Pro, Model Category=Proprietary2026.0155.44
CoTModel=GPT-5, Model Category=Proprietary2026.0154.71
CoTModel=Gemini 2.5 Pro, Model Category=Proprietary2026.0153.91
CoTModel=Qwen3-30B, Model Category=Open Source2026.0153.35
F-1Model=Qwen3-30B, Model Category=Open Source2026.0151.82
Zero-ShotModel=Qwen3-30B, Model Category=Open Source2026.0150.99
PoTModel=Gemini 2.5 Pro, Model Category=Proprietary2026.0149.06
PoTModel=GPT-5, Model Category=Proprietary2026.0147.73
SeirênesBackbone=Qwen3-30B-A3B, same budget=true2026.0546.7
DAPOBackbone=Qwen3-30B-A3B, same budget=true2026.0544.3
PoTModel=Qwen3-30B, Model Category=Open Source2026.0142.61
SeirênesBackbone=Qwen3-4B-Instruct, same budget=true2026.0541.9
Qwen3-30B-A3BBackbone=Qwen3-30B-A3B2026.0540.5
InTBackbone=Qwen3-4B-Instruct2026.0538.8
Dr.GRPOBackbone=Qwen3-4B-Instruct, same budget=true2026.0538
DAPOBackbone=Qwen3-4B-Instruct, same budget=true2026.0537.7
Qwen3-4B-InstructBackbone=Qwen3-4B-Instruct2026.0534.1
Scaf-GRPOBackbone=Qwen3-4B-Instruct2026.0533.9
SAGEBackbone=Qwen3-4B-Instruct2026.0532.5
Zero-ShotModel=DeepSeek-V3.1, Model Category=Open Source2026.0130.54
F-1Model=DeepSeek-V3.1, Model Category=Open Source2026.0127.64
CoTModel=DeepSeek-V3.1, Model Category=Open Source2026.0127.56
PoTModel=Qwen3-235B, Model Category=Open Source2026.0123.94
Zero-ShotModel=Qwen3-235B, Model Category=Open Source2026.0121.85
F-1Model=Qwen3-235B, Model Category=Open Source2026.0121.84
CoTModel=Qwen3-235B, Model Category=Open Source2026.0120.45
PoTModel=DeepSeek-V3.1, Model Category=Open Source2026.0117.48
SeirênesBackbone=Qwen2.5-7B-Instruct, same budget=true2026.0513.3
SCRLBase Model=Qwen3-14B-Base2026.0511.9
SAGEBackbone=Qwen2.5-7B-Instruct2026.0511.2
LUFFYBackbone=Qwen2.5-7B-Instruct2026.0511
SFTBase Model=Qwen3-14B-Base2026.0511
QuestABase Model=Qwen3-14B-Base2026.0510.4
ExGRPOBackbone=Qwen2.5-7B-Instruct2026.059.9
GRPOBase Model=Qwen3-14B-Base2026.059.9
DAPOBase Model=Qwen3-14B-Base2026.059.9
DAPOBackbone=Qwen2.5-7B-Instruct, same budget=true2026.059.5
NuRLBase Model=Qwen3-14B-Base2026.059.4
Qwen2.5-7B-InstructBackbone=Qwen2.5-7B-Instruct2026.058.8
SCRLBase Model=Qwen3-4B-Base2026.058.7
DAPOBase Model=Qwen3-4B-Base2026.058.3
QuestABase Model=Qwen3-4B-Base2026.058.3
GRPOBase Model=Qwen3-4B-Base2026.057.8
Initial ModelBase Model=Qwen3-14B-Base2026.057.4
SFTBase Model=Qwen3-4B-Base2026.057.1
NuRLBase Model=Qwen3-4B-Base2026.057.1
SCRLBase Model=Llama3.2-3B-Instruct2026.056.8
QuestABase Model=Llama3.2-3B-Instruct2026.056.1
Initial ModelBase Model=Qwen3-4B-Base2026.056.1
NuRLBase Model=Llama3.2-3B-Instruct2026.055.9
DAPOBase Model=Llama3.2-3B-Instruct2026.055.8
GRPOBase Model=Llama3.2-3B-Instruct2026.055.4
Initial ModelBase Model=Llama3.2-3B-Instruct2026.054.5
SFTBase Model=Llama3.2-3B-Instruct2026.052.6