Described embodiments provide systems and methods for latency optimization for cloud applications. An agent of a client device comprising an audio decoder and a video decoder can monitor video and audio data paths of an application communicating audio/video (A/V) data from one or more servers to the client device. The agent can measure, using the audio decoder and the video decoder, an A/V latency and a lip-sync status of the video and audio data paths of the application. The agent can determine, based on at least one or more measurements of the A/V latency and the lip-sync status, to enable a low latency mode for at least one of the video decoder or the audio decoder. The agent can configure, responsive to the determination, the low latency mode on one of the video decoder or the audio decoder.