Waver 1.0 is a breakthrough unified generative video model developed by ByteDance, designed to support multi-mode content creation — including text-to-video (T2V), image-to-video (I2V) and text-to-image (T2I). The platform offers high resolution (up to 1080p), smooth motion, multi-shot narrative capability, rich stylistic control and fast generation. Users can input natural language prompts such as “a dragon flying over a medieval castle at sunset” or upload an existing image and convert it into a dynamic video. Waver 1.0 excels at coherent visual storytelling across multiple scenes, maintains consistency of characters and style, and supports diverse aesthetics from photorealism to stylised illustration. The model is designed for creators, marketers, agencies and storytellers who need cinematic-level visuals without traditional production costs. It operates on a credit-based or freemium model and is available via waitlist or API.