ApplicationsFeaturedUpdated 2026-10-24
Text-to-Video
Generating video clips from text descriptions.
#文生视频#视频生成#AI 视频
In plain language
Text-to-video models generate visuals, motion and camera work from prompts. Controls include subject, action, camera movement, lighting and duration. Current models still struggle with physical consistency, long videos and precise control.
Example
Prompt: rainy neon street, slow push-in, cinematic.
Recommended tools
Related terms
Models & ArchitectureFeatured
Diffusion Model
A generative model that turns random noise into images, video or audio through iterative denoising.
3 related terms
Applications
Image-to-Video
Generating motion video starting from a still image.
3 related terms
Models & Architecture
World Model
An internal representation that learns how an environment works and predicts the consequences of actions.
3 related terms
Engineering
Agent Memory
A mechanism for an agent to store and retrieve user preferences, past tasks and key facts across sessions.
3 related terms