List your company

Mathematical Reasoning on Mathematical Reasoning Suite Overall

63.9Average ScoreSeirênes
11.79625.32338.8552.377May 12, 2026

Evaluation results

MethodLinks
SeirênesBackbone=Qwen3-30B-A3B, same budget=true2026.0563.97.2
DAPOBackbone=Qwen3-30B-A3B, same budget=true2026.0560.13.4
SeirênesBackbone=Qwen3-4B-Instruct, same budget=true2026.055810.2
Qwen3-30B-A3BBackbone=Qwen3-30B-A3B2026.0556.7—
Dr.GRPOBackbone=Qwen3-4B-Instruct, same budget=true2026.0553.75.9
InTBackbone=Qwen3-4B-Instruct2026.0553.35.5
DAPOBackbone=Qwen3-4B-Instruct, same budget=true2026.0552.14.3
SAGEBackbone=Qwen3-4B-Instruct2026.0548.20.4
Scaf-GRPOBackbone=Qwen3-4B-Instruct2026.0548.10.3
Qwen3-4B-InstructBackbone=Qwen3-4B-Instruct2026.0547.8—
SeirênesBackbone=Qwen2.5-7B-Instruct, same budget=true2026.0522.99.1
LUFFYBackbone=Qwen2.5-7B-Instruct2026.0518.74.9
SAGEBackbone=Qwen2.5-7B-Instruct2026.0518.64.8
ExGRPOBackbone=Qwen2.5-7B-Instruct2026.0517.73.9
DAPOBackbone=Qwen2.5-7B-Instruct, same budget=true2026.0517.43.6
Qwen2.5-7B-InstructBackbone=Qwen2.5-7B-Instruct2026.0513.8—