<p>The CMIP6 model incorporates various socio-economic and technological development scenarios and provides a comprehensive assessment of climate change. It differs from earlier versions of climate models in terms of the number of modeling groups, future scenarios evaluated, and experiments performed. Selecting a suitable subset of CMIP6 models poses a significant challenge for climate change projections. Hence, this study aims to evaluate how well CMIP6 models simulate rainfall over the Didessa, Southwest of Ethiopia. The performance of the model was tested using observed data from 29&#xa0;years (1985 to 2014). The statistical and Taylor diagram approaches are used to check the model’s performance. The results showed that the model’s performance was affected by the seasons. The GFDL-ESM4, GFDL-CM4, and CMCC-ESM2 models and their ensembles of them performed better on an annual basis. The GFDL-ESM4, GFDL-CM4, and MRI-ESM2-0 models were best performed in the summer season (June, July, and August). GFDL-CM4 and GFDL-ESM4 performed admirably during the spring season (September, October, and November). GFDL-ESM4, GFDL-ESM4, MPI-ESM1-2-HR, and MRI-ESM2-0 fit well with the observed data in the Autumn season (March, April, and May). In the dry season (December, January, and February), the CESM2-WACCM model outperformed in modeling rainfall. The MIROC6 overestimated the rainfall in all months. All models were overestimated in wet months and underestimated in dry months. The spatial distribution of the rainfall shows that all models simulated peak rainfall in central parts of the watershed and low rainfall values downstream of the sub-basin.</p>