context.vn

mmmlu

6 models evaluated

#ModelProviderTypeScore
1Interfaze BetaInterfazeClosedinterfaze-beta
2Qwen3.7 MaxAlibabaClosedqwen3-7-max
3DeepSeek V4 Pro BaseDeepSeekOpendeepseek-v4-pro-base
4Qwen3.7 PlusAlibaba · Closed89.0%
5DeepSeek V4 Flash BaseDeepSeek · Open weight88.8%
6Gemma 4 12BGoogle · Open weight83.4%