Abstract: Methods, apparatus, and processor-readable storage media for determining a virtual machine configuration based on application source code are provided herein. An example computer-implemented method includes parsing source code of an application to determine one or more features of the application; providing the one or more features to at least one machine learning model, wherein the machine learning model is trained based at least in part on historical usage data associated with one or more virtual machines configured for one or more other applications; obtaining, from the at least one machine learning model, one of a plurality of virtual machine configurations for the application; and initiating a configuration of at least one virtual machine for the application based at least in part on the virtual machine configuration obtained from the at least one machine learning model.