SuperCLUE has released the August rankings for Chinese large models, which include the overall leaderboard, the OPEN multi-round open-ended questions leaderboard, the OPT three major abilities objective questions leaderboard, the top ten basic abilities leaderboard, and the open-source leaderboard. The evaluation selected 16 general-purpose large language models and used 3,337 brand-new test questions. This evaluation found that the performance of domestic large models on Chinese tasks is continuously narrowing the gap with GPT3.5.