Weeds are recognized as the most aggressive, troublesome, and competitive elements within croplands. Climate change may affect the geographical distribution of existing weeds or the invasion of weeds in new areas. Therefore, in the current study, modelling was carried out to explore and predict the invasion potential of Trianthema portulacastrum and Ageratum conyzoides in India under current as well as future climatic conditions. Future climatic scenarios under Representative Concentration Pathways (RCPs) 4.5 and 8.5 for the years 2050 and 2070 were considered and modelling was performed using regression techniques and machine learning approaches with Ensemble technique. Mutually least correlated eight bioclimatic variables along with soil and elevation data were used for the modelling over 375 and 379 occurrence locations of the T. portulacastrum and A. conyzoides, respectively. True Skill Statistic (TSS) was used to evaluate the models’ predictive performance, while AUC of the ROC and Kappa were used to crosscheck their performance. Results revealed that Ensemble model outperformed the individual models for both the species with higher AUC, TSS and Kappa values. Bioclimatic variables such as temperature seasonality, annual mean temperature, and minimum temperature of the coldest month were found to govern the potential distribution of both the species. Results of potential distribution were obtained in four climate suitability classes: not suitable (0-0.2), low (0.2-0.4), moderately (0.4-0.6), and highly suitable (>0.6). Modelling suggests the expansion of suitable ranges of T. portulacastrum in India under future climatic scenarios, whilst A. conyzoides are expected to predominantly contract in the future.