context.vn

mmmu Pro Python

9 models evaluated

#ModelProviderTypeScore
1GPT-5.6 SolOpenAIClosedgpt-5-6-sol
2Kimi K3Moonshot AIClosedkimi-3
3GPT-5.5OpenAIClosedgpt-5-5
4GPT-5.4OpenAI · Closed82.1%
5GPT-5.6 TerraOpenAI · Closed82%
6Kimi K2.6Moonshot AI · Open weight80.1%
7GPT-5.6 LunaOpenAI · Closed79.5%
8GPT-5.4 miniOpenAI · Closed78%
9GPT-5.4 nanoOpenAI · Closed69.5%