List your company
ResearchFormerly WizWand

Mathematical Reasoning on AIME 24 (pass@2, pass@4, pass@8, pass@16)

27.03Pass@2Qwen3-235B-A22B-Instr.
-0.64446.540313.72520.9097Mar 30, 2026
Updated 7mo ago

Evaluation results

MethodLinks
Qwen3-235B-A22B-Instr.Parameter Scale=235B, Training Strategy=Commercial Baseline2026.0327.0328.323033.33
DeepSeek-R1-671B-0528Parameter Scale=671B, Training Strategy=Commercial Baseline2026.0321.0328.0732.1433.33
Qwen2.5-7B-ERPOParameter Scale=7B, Training Strategy=ERPO2026.0317.8922.4527.3833.33
Qwen2.5-7B-GRPOParameter Scale=7B, Training Strategy=GRPO2026.0316.2821.4927.0933.33
Qwen2.5-3B-ERPOParameter Scale=3B, Training Strategy=ERPO2026.0311.2216.4323.433.33
Qwen2.5-3B-GRPOParameter Scale=3B, Training Strategy=GRPO2026.038.5812.8118.4426.67
Qwen2.5-1.5B-ERPOParameter Scale=1.5B, Training Strategy=ERPO2026.036.8111.417.0823.33
Qwen2.5-1.5B-GRPOParameter Scale=1.5B, Training Strategy=GRPO2026.036.1910.0615.5623.33
Qwen2.5-7B-BaseParameter Scale=7B, Training Strategy=Base2026.036.1910.0615.5623.33
Qwen2.5-3B-BaseParameter Scale=3B, Training Strategy=Base2026.033.926.9511.2116.67
Qwen2.5-7B-SFTParameter Scale=7B, Training Strategy=SFT2026.032.865.510.1116.67
Qwen2.5-1.5B-SFTParameter Scale=1.5B, Training Strategy=SFT2026.031.673.336.6713.33
Qwen2.5-3B-SFTParameter Scale=3B, Training Strategy=SFT2026.031.643.175.8910
Qwen2.5-1.5B-BaseParameter Scale=1.5B, Training Strategy=Base2026.030.420.831.673.33