The Yunnan–Guizhou quasi-stationary front (YGQSF) is a critical weather system in Southwest China during the cold season (November to April), frequently introducing low-temperature disasters. Accurate forecasting of YGQSF poses significant challenges for operational models. This study evaluates the performance of ECMWF Integrated Forecasting System (IFS) in forecasting YGQSFs by utilizing hourly observational data and a linear fitting method for frontal line identification. From 2020 to 2023, a total of 392 and 261 YGQSFs were identified in observational data and the IFS forecast, respectively. From east to west, the frontal lines gradually change from northwest–southeast-oriented to quasi-north–south-oriented. The northern segments of the frontal lines are concentrated east of the Hengduan Mountains, while the southern segments are dispersed across four high-occurrence regions on the Yungui Plateau. These high-occurrence regions are located on local highlands or their eastern slopes, highlighting the influence of topographic blocking effects. These spatial characteristics of the YGQSFs are well captured by the IFS, especially for the high-occurrence region between 103.5°E and 105°E, which is related to the pronounced spatial temperature gradients enhanced by the large elevation difference. Further analysis focuses on the impact of spatial temperature gradients on the YGQSF forecasting, particularly examining Hits, Misses, and False alarms. The model demonstrates superior performance in forecasting YGQSFs characterized by large temperature gradients and substantial cold air accumulation. Conversely, reduced temperature gradients, stemming from weaker cold air to the east or weaker warm air to the west, increase the risk of Misses. Temporally, the missed YGQSFs are uniformly distributed, while spatially they are scattered. False alarms, however, peak between February and March and are concentrated between 103.5°E and 107.25°E. Overall, forecasting YGQSFs with weak intensity, short frontal lines, or meridional orientation remains particularly challenging. The biases revealed in this study provide valuable insights for enhancing operational forecasting accuracy.

Evaluation constitutes a crucial component in the creation and application of numerical models. As these models have escalated in complexity in recent years, there has been an increasing demand for more accurate forecasting, thereby catalyzing substantial advancements in model verification and evaluation techniques. This paper begins with a review of recent domestic and international progress in model evaluation and verification, and subsequently delves into the significant contributions made by Chinese researchers in evaluating the dynamical frameworks of next-gene-ration numerical models, the spatiotemporal variability of the East Asian monsoon, and the assessment of weather and climate conditions around the Qinghai–Xizang Plateau (QXP). Given the identification of precipitation as a key process, the paper presents a comprehensive review of evaluation efforts pertaining to diurnal variations and process evolution in precipitation. The paper concludes by outlining potential future research outlooks, including the integration of multi-source observation data, the development of evaluation techniques specific to certain models and application contexts, and the synergy between model evaluation and machine learning technology.