The issue of how efficiently farmers are using various farm inputs in crop cultivation has been an important topic of research over the years. Many studies from India seem to suggest that the farm inputs are mostly used sub-optimally. However, not many studies are available on the inputs use efficiency covering different crops and states using recent temporal data. In this study, therefore, an attempt has been made to find out the efficiency of different inputs used for cultivating six different crops, namely, paddy, wheat, gram, groundnut, cotton and sugarcane by utilising cost of cultivation survey data from 1985-86 to 2013-14. The study shows that the value of crop output (at 1986-87 prices) per unit of input generated interms of rupees has not increased consistently over the years in both foodgrain and non-foodgrain crops. The efficiency of yield augmenting inputs such as fertilisers and irrigation either has declined or fluctuated in most crops and states. There is no conclusive evidence to show that the inputs are used more efficiently in high productivity states than that of the low productivity states in all the six crops considered for the analysis. In crops like gram and sugarcane, the low productivity states have outperformed the high productivity states not only in the overall resource use efficiency but even at the individual level input use efficiency. The regression analysis carried out to find out the efficiency of different inputs over time seems to suggest that the inputs have not been used efficiently in all the six crops. Even the yield augmenting cost-intensive inputs such as fertilisers, irrigation and seed seem to have not been used efficiently over time. This pattern is observed in both the high and low productivity states in all the six crops considered for the analysis. The study concludes that the sub-optimal price received by the farmers in the market may have dampened the efficiency of inputs used for crops cultivation. © 2017 Indian Society of Agricultural Economics. All rights reserved.