Wan

Wan is an AI-driven creative platform designed to generate and edit visual media including images and videos. It offers a wide range of tools that allow users to create content from scratch or modify existing media using text prompts, reference images, and video inputs. The core functionality revolves around multiple generation and editing modes, each tailored to specific creative needs. For image generation, Wan supports text-to-image conversion with an emphasis on accurate text rendering. Users can input descriptive text and the system produces corresponding images that include legible long-form text, such as slogans, signs, or book covers. This feature is particularly useful for graphic designers who need to visualize concepts with embedded typography, or for marketers creating social media visuals with specific wording. Additionally, the platform handles sequential image generation, where multiple images maintain consistent visual elements like character designs, color palettes, or scene compositions. This enables storyboarding for animations, comic strips, or product catalogues where style coherence is critical. Video generation is another major capability. Users can convert text descriptions directly into video clips, or provide a reference video along with a text prompt to produce new footage that mimics the motion, camera angles, and scene dynamics of the original. This “creative video transfer” feature allows filmmakers and content creators to replicate complex actions, such as a dancer’s movements or a drone’s panning shot, without needing to reshoot. The platform also supports instruction-based video editing: by giving natural language commands like “change the background to a forest” or “replace the character’s jacket with a blue coat,” users can modify spec