1. Introduction Intercropping offers the prospect of providing greater and more stable yield than sole-cropping in the face of climate change and increased climate variability. Cereal-legume intercropping is common in tropical cropping systems, but often with low legume density and limited or no nutrient inputs. Combining intercropping with integrated soil fertility management is a solution for promoting sustainable intensification. Long-term experiments investigating the impact of variations in soil and climate on intercrop performances are scarce throughout sub-Saharan Africa. Several models were designed to simulate intercropping, however their robustness and accuracy in reproducing intercropping functioning and performance have never been extensively evaluated for a range of tropical environments. This work aims to evaluate the robustness and accuracy of the STICS soil-crop model in simulating productivity and crop interactions for contrasting cereal-legume intercrops in tropical conditions. 2. Materials and Methods The STICS model (Beaudoin et al., 2023) was tested using data collected in on-farm and on-station experiments in Mali, Senegal, Burkina Faso, and Brazil. Various combinations of cereals (maize, sorghum, and millet) and legumes (pigeon pea and cowpea) were compared in sole cropping and intercropping. The experiments included contrasting spatial patterns, fertilizer inputs, crop varieties, sowing dates, and cereal/legume densities. The model was calibrated on sole cropping, and evaluated on intercropping. Observed and simulated partial Land Equivalent Ratio (pLER) was calculated to evaluate the ability of the model in simulating competition and complementarity effects in intercropping. 3. Results We found that the STICS model had similar accuracy when simulating cereal grain yield of sole crops and intercrops. Model accuracy in simulating legumes was lower with intercropping compared with sole cropping. The model correctly reproduced competition and co