List your company

STEM Knowledge on MMLU-Pro

84.3AccuracyDeepSeek V3.2
81.38882.14482.983.656Dec 30, 2025

Evaluation results

MethodLinks
DeepSeek V3.2Evaluation Mode=Chat2025.1284.3
LongCat-Flash Exp-ChatEvaluation Mode=Chat2025.1284
LongCat-Flash ChatEvaluation Mode=Chat2025.1282.7
GLM 4.6Evaluation Mode=Chat2025.1281.5