We develop Video Latent Diffusion Models (Video LDMs) for computationally efficient high-resolution video synthesis. We first pre-train an LDM on images only; then, we turn the image generator into a video generator by introducing a temporal dimension to the latent space diffusion model and fine-tuning on encoded image sequences, i.e., videos. We demonstrate high-resolution text-to-video synthesis and in-the-wild driving video generation.