This study used the ECOSSE model (v. 5.0.1) to simulate soil respiration (Rs) fluxes estimated from ecosystem respiration (Reco) for eight European permanent grassland (PG) sites with varying grass species, soils, and management. The main aim was to evaluate the strengths and weaknesses of the model in estimating Rs from grasslands, and to gain a better understanding of the terrestrial carbon cycle and how Rs is affected by natural and anthropogenic drivers. Results revealed that the current version of the ECOSSE model might not be reliable for estimating daily Rs fluxes, particularly in dry sites. The daily estimated and simulated Rs ranged from 0.95 to 3.1 g CO2-C m−2, and from 0.72 to 1.58 g CO2-C m−2, respectively. However, ECOSSE could still be a valuable tool for predicting cumulative Rs from PG. The overall annual relative deviation (RD) value between the cumulative estimated and simulated annual Rs was 11.9%. Additionally, the model demonstrated accurate simulation of Rs in response to grass cutting and slurry application practices. The sensitivity analyses and attribution tests revealed that increased soil organic carbon (SOC), soil pH, temperature, reduced precipitation, and lower water table (WT) depth could lead to increased Rs from soils. The variability of Rs fluxes across sites and years was attributed to climate, weather, soil properties, and management practices. The study suggests the need for additional development and application of the ECOSSE model, specifically in dry and low input sites, to evaluate the impacts of various land management interventions on carbon sequestration and emissions in PG.