This study designed and validated an artificial intelligence-based evaluation model at Payam Noor University. The research type was a sequential mixed method with an exploratory approach in terms of data and included two qualitative and quantitative parts. In the qualitative part, the meta-synthesis and Delphi stages were used. The statistical population in the meta-synthesis stage included theoretical foundations and related backgrounds from domestic and foreign databases, and in the Delphi stage, it included 15 experts using a purposive non-random sampling method. In the quantitative part, the statistical population included all professors at Payam Noor University, 245 of whom were selected using a cluster sampling method. Data were collected in the qualitative part through a systematic literature review and in the Delphi stage with a worksheet. In the quantitative part, a 75-item questionnaire extracted from the results of the qualitative part was used. The validity and reliability of the tools were examined, and the results indicated their appropriate validity. Data analysis in the qualitative part was performed using systematic analysis and Kendall's coefficient of agreement, and in the quantitative part, descriptive and inferential statistics (confirmatory factor analysis) were performed using Maxqda, SPSS, and Smart PLS software. The findings showed that the evaluation model includes the dimensions of design, implementation, analysis of results, and improvement and development. This model was presented and its validity was confirmed based on the aforementioned dimensions and components.