Recent advancements in CMIP6 Global Climate Models (GCMs), guided by Shared Socioeconomic Pathways under the IPCC’s 6th Assessment Report, have significantly enhanced the simulation of temperature-related climatic variables. This study assesses the performance of 24 CMIP6 GCMs in reproducing minimum and maximum temperatures for South Goa, India, a climatically sensitive region situated on the western coastline of India and strongly influenced by the Indian summer monsoon. A multi-model ensemble evaluation is performed using Compromise Programming, integrating four key statistical metrics—R2 (Coefficient of Determination), Percentage Bias (PBIAS), Normalized Root Mean Square Error (NRMSE), and Nash–Sutcliffe Efficiency (NSE)—into a unified model-ranking framework. Through comparison with observed temperature data, the study identifies CanESM5 as the most reliable GCM for simulating both minimum and maximum temperatures in the coastline region. This approach, leveraging a composite performance index, advances beyond conventional single-metric evaluations by offering enhanced precision and objectivity. The findings not only support model selection for localized climate impact assessments but also underscore the utility of Compromise Programming as a robust and replicable tool for validating GCM performance in monsoon driven coastal environments.