List your company
ResearchFormerly WizWand

Mathematical Reasoning on AIME 2024 (Score Metric)

93.3ScoreQwen3-30B-A3B-Thinking-2507
49.93261.19172.4583.709Sep 25, 2025
Updated 8mo ago

Evaluation results

MethodLinks
Qwen3-30B-A3B-Thinking-2507Sampling=Bo∞ (Best-of-infinity sampling)2025.0993.3
Qwen3-30B-A3B-Thinking-2507Sampling=Bo1 (Best-of-1 sampling)2025.0990.5
AM-Thinking-v1Sampling=Bo∞ (Best-of-infinity sampling)2025.0990
GPT-OSS-20BSampling=Bo∞ (Best-of-infinity sampling), reasoning mode=medium2025.0990
EXAONE-Deep-32BSampling=Bo∞ (Best-of-infinity sampling)2025.0986.7
MetaStone-S1-32BSampling=Bo∞ (Best-of-infinity sampling)2025.0986.7
NVIDIA-Nemotron-Nano-9B-v2Sampling=Bo∞ (Best-of-infinity sampling)2025.0986.7
Phi-4-reasoningSampling=Bo∞ (Best-of-infinity sampling)2025.0986.7
Qwen3-14BSampling=Bo∞ (Best-of-infinity sampling)2025.0986.7
Qwen3-14BSampling=Bo1 (Best-of-1 sampling)2025.0983
MetaStone-S1-32BSampling=Bo1 (Best-of-1 sampling)2025.0982
LIMO-v2Sampling=Bo∞ (Best-of-infinity sampling)2025.0980
Qwen3-4BSampling=Bo∞ (Best-of-infinity sampling)2025.0980
AM-Thinking-v1Sampling=Bo1 (Best-of-1 sampling)2025.0978.9
GPT-OSS-20BSampling=Bo1 (Best-of-1 sampling), reasoning mode=medium2025.0978
Qwen3-4BSampling=Bo1 (Best-of-1 sampling)2025.0973.5
Datarus-R1-14B-previewSampling=Bo∞ (Best-of-infinity sampling)2025.0973.3
Phi-4-reasoningSampling=Bo1 (Best-of-1 sampling)2025.0972.9
NVIDIA-Nemotron-Nano-9B-v2Sampling=Bo1 (Best-of-1 sampling)2025.0971.6
EXAONE-Deep-32BSampling=Bo1 (Best-of-1 sampling)2025.0971.5
LIMO-v2Sampling=Bo1 (Best-of-1 sampling)2025.0962
Datarus-R1-14B-previewSampling=Bo1 (Best-of-1 sampling)2025.0951.6