List your company
ResearchFormerly WizWand

Mathematical Reasoning on AIME 2026 (held-out)

76.1AccuracyAntiSD
14.63630.59346.5562.507May 12, 2026
Updated 5mo ago

Evaluation results

MethodLinks
AntiSDBackbone=Olmo3-7B-TK, Training Condition=+AntiSD, Speedup=2.0×, Peak-mean step=40, Sampling strategy=avg@322026.0576.1
GRPOBackbone=Olmo3-7B-TK, Training Condition=+GRPO, Speedup=1.0×, Peak-mean step=80, Sampling strategy=avg@322026.0575.3
AntiSDBackbone=Qwen3-4B-IT-2507, Training Condition=+AntiSD, Speedup=10.0×, Peak-mean step=100, Sampling strategy=avg@322026.0574.4
AntiSDBackbone=Qwen3-30B-A3B, Training Condition=+AntiSD, Speedup=2.9×, Peak-mean step=140, Sampling strategy=avg@322026.0574.1
SDBackbone=Olmo3-7B-TK, Training Condition=+SD, Peak-mean step=40, Sampling strategy=avg@322026.0574
AntiSDBackbone=Qwen3-8B, Training Condition=+AntiSD, Speedup=5.0×, Peak-mean step=180, Sampling strategy=avg@322026.0573.7
Olmo3-7B-TKBackbone=Olmo3-7B-TK, Training Condition=Base, Sampling strategy=avg@322026.0573.3
GRPOBackbone=Qwen3-30B-A3B, Training Condition=+GRPO, Speedup=1.0×, Peak-mean step=150, Sampling strategy=avg@322026.0565.5
GRPOBackbone=Qwen3-8B, Training Condition=+GRPO, Speedup=1.0×, Peak-mean step=200, Sampling strategy=avg@322026.0564.2
GRPOBackbone=Qwen3-4B-IT-2507, Training Condition=+GRPO, Speedup=1.0×, Peak-mean step=200, Sampling strategy=avg@322026.0563.5
AntiSDBackbone=Olmo3-7B-IT, Training Condition=+AntiSD, Speedup=9.5×, Peak-mean step=200, Sampling strategy=avg@322026.0555.2
Qwen3-4B-IT-2507Backbone=Qwen3-4B-IT-2507, Training Condition=Base, Sampling strategy=avg@322026.0553.8
GRPOBackbone=Olmo3-7B-IT, Training Condition=+GRPO, Speedup=1.0×, Peak-mean step=190, Sampling strategy=avg@322026.0552.1
SDBackbone=Qwen3-4B-IT-2507, Training Condition=+SD, Peak-mean step=10, Sampling strategy=avg@322026.0552
SDBackbone=Olmo3-7B-IT, Training Condition=+SD, Peak-mean step=10, Sampling strategy=avg@322026.0546.6
Olmo3-7B-ITBackbone=Olmo3-7B-IT, Training Condition=Base, Sampling strategy=avg@322026.0544.4
SDBackbone=Qwen3-30B-A3B, Training Condition=+SD, Peak-mean step=60, Sampling strategy=avg@322026.0534
SDBackbone=Qwen3-8B, Training Condition=+SD, Peak-mean step=200, Sampling strategy=avg@322026.0526.9
Qwen3-30B-A3BBackbone=Qwen3-30B-A3B, Training Condition=Base, Sampling strategy=avg@322026.0521.8
Qwen3-8BBackbone=Qwen3-8B, Training Condition=Base, Sampling strategy=avg@322026.0517