A method, performed by a server, of streaming image content, including obtaining one or more frames of the image content which is currently being executed by an electronic device; identifying a plurality of possible user inputs corresponding to the one or more frames; generating a plurality of prediction frame sets respectively corresponding to the plurality of possible user inputs, wherein each prediction frame set of the plurality of prediction frame sets includes a plurality of prediction frames to be displayed after the one or more frames; obtaining metadata from the one or more frames and the plurality of prediction frame sets; generating streaming data of the image content to be provided to the electronic device, based on at least one the plurality of prediction frame sets or the metadata; and transmitting the generated streaming data to the electronic device.