List your company
ResearchFormerly WizWand

Mathematical Reasoning on HMMT Feb25

94.6Pass@1Nemotron-Cascade-2 30B-A3B
7.052829.781452.5175.2386Jul 17, 2025Sep 4, 2025Oct 23, 2025Dec 11, 2025Jan 29, 2026Mar 19, 2026May 8, 2026
Updated 5mo ago

Evaluation results

MethodLinks
Nemotron-Cascade-2 30B-A3B2026.0394.6
Nemotron-3-Super 120B-A12B2026.0393.7
Qwen3.5 35B-A3B2026.0389
Nemotron-3-Nano 30B-A3BOfficial/Recommended Settings=true2026.0384.6
Qwen3-8Bk (responses per question)=32, sampling temperature=0.7, top-p=0.952025.0744.79
GPT-5.2-chat (teacher)2026.0543.75
QUESTA-Nemotron-1.5Bk (responses per question)=32, sampling temperature=0.7, top-p=0.952025.0741.67
ROPDThinking Mode=Thinking2026.0539.16
T-JudgeThinking Mode=Thinking2026.0538.75
OVDThinking Mode=Thinking2026.0538.34
GADThinking Mode=Thinking2026.0535
Qwen3-4B (student)Thinking Mode=Thinking2026.0533.33
Qwen3-30B-A3B (teacher)Access=–2026.0533.33
DeepSeek-R1-Distill-32Bk (responses per question)=32, sampling temperature=0.7, top-p=0.952025.0733
ROPDThinking Mode=Non-Thinking2026.0531.69
Nemotron-1.5Bk (responses per question)=32, sampling temperature=0.7, top-p=0.952025.0731.56
OVDThinking Mode=Non-Thinking2026.0529.11
T-JudgeThinking Mode=Non-Thinking2026.0528.94
ROPDAccess=text2026.0525.4
ExOPDAccess=logit2026.0522.42
Qwen3-1.7Bk (responses per question)=32, sampling temperature=0.7, top-p=0.952025.0722.19
LOPDAccess=logit2026.0520.42
GADThinking Mode=Non-Thinking2026.0512.84
DeepSeek-R1-Distill-1.5Bk (responses per question)=32, sampling temperature=0.7, top-p=0.952025.0712
SFTAccess=text2026.0511.62
Qwen3-4B (student)Thinking Mode=Non-Thinking2026.0510.42
Qwen3-4B (student)Access=–2026.0510.42