Modalities
Text-to-Video
AI that generates video clips from a text description — the next frontier after text-to-image, with rapidly improving quality.
01 ——
In plain English
Text-to-video is a class of generative AI that produces video clips from a written prompt. It's an extension of text-to-image into the time dimension — significantly harder, but advancing fast.
Leading text-to-video tools:
- OpenAI Sora — long, photorealistic clips
- Runway Gen-3 / Gen-4 — popular for filmmakers
- Pika — short clips, easy to use
- Luma Dream Machine — quality on par with Runway
- Google Veo — high-resolution, high-fidelity
Current limits:
- Length — most tools cap at 5–30 seconds per clip
- Consistency — characters and objects can morph between frames
- Physics — bodies can warp, objects can pass through each other
- Cost — video generation is much more compute-intensive than images
Despite the limits, text-to-video is already used in advertising, music videos, and short-form social content.
02 ——
Related terms
Text-to-Image
AI that generates new images from a written description — the technology behind tools like Midjourney, DALL-E, and Stable Diffusion.
Diffusion Model
The type of AI model behind most modern image and video generators — it learns to create content by reversing a noising process.
Generative AI
AI systems that create new content — text, images, audio, video, or code — rather than just classifying or predicting from existing data.
Multi-modal
An AI model that can understand and work with multiple types of input — text, images, audio, or video — not just text.
Back to glossaryLast reviewed May 2026