List your company
ResearchFormerly WizWand

Mathematical Reasoning on HMMT (Solved Rate %)

80.01Solved RateGPT-OSS-120B
-3.1426818.44503540.0327561.620465May 21, 2025Jul 14, 2025Sep 6, 2025Oct 30, 2025Dec 23, 2025Feb 15, 2026Apr 10, 2026
Updated 5mo ago

Evaluation results

MethodLinks
GPT-OSS-120BSampling Strategy=4-sample mean2026.0480.01
Aryabhata 2Sampling Strategy=4-sample mean2026.0478.96
GPT-OSS-20BSampling Strategy=4-sample mean2026.0477.42
GPT-5 MiniSampling Strategy=4-sample mean2026.0470.97
Nemotron 3 Nano 30B A3BSampling Strategy=4-sample mean2026.0465.86
GPT-5 NanoSampling Strategy=4-sample mean2026.0463.98
Gemini 2.5 FlashSampling Strategy=4-sample mean2026.0459.13
Qwen3-30B-A3B (Thinking)Sampling Strategy=4-sample mean2026.0451.88
ReflexionModel=Qwen3 32B think, Context Window=32k2025.050.6
ICRLModel=Qwen3 32B think, Context Window=32k2025.050.6
Self-RefineModel=Qwen3 32B think, Context Window=32k2025.050.5666
BaseModel=Qwen3 32B think, Context Window=32k2025.050.52
ICRLModel=Qwen3 32B, Context Window=32k2025.050.3333
ReflexionModel=Qwen3 32B, Context Window=32k2025.050.2333
ICRLModel=Llama 4 Maverick, Context Window=32k2025.050.2
Self-RefineModel=Qwen3 32B, Context Window=32k2025.050.1666
Self-RefineModel=Llama 4 Maverick, Context Window=32k2025.050.1333
Self-RefineModel=Phi-4, Context Window=16k2025.050.1333
ReflexionModel=Phi-4, Context Window=16k2025.050.1333
ICRLModel=Phi-4, Context Window=16k2025.050.1333
ReflexionModel=Llama 4 Maverick, Context Window=32k2025.050.1
BaseModel=Qwen3 32B, Context Window=32k2025.050.0914
BaseModel=Llama 4 Maverick, Context Window=32k2025.050.085
BaseModel=Phi-4, Context Window=16k2025.050.0555