Text to Video AI: make a video from a prompt
Describe a scene and get a video. Veo 3.1, Kling 3.0, Seedance 2.0, MiniMax H3, Grok Imagine and Lumyn's own video models in one browser studio.
Text to video turns a written description into a short clip. Describe the subject, the action, the setting and the camera: "a red vintage car drives along a coastal road at sunset, drone shot following from behind". The model decides everything you do not specify, so the more you care about something, the more you should say about it.
Models to try
- Veo 3.1 Video: Google's model, cinematic and detailed.
- Kling 3.0 Pro Video: realistic people and motion.
- Seedance 2.0 Video: good prompt following and multi-shot scenes.
- MiniMax H3 Video: dynamic action.
- Grok Imagine Video: xAI's video model.
- Lumyn 2.6 Video: the house model, uncensored.
Compare them all with prices on Models.
Prompt tips
Put the camera in the prompt ("close-up", "slow dolly in", "handheld"). Keep one main action per clip. If a character has to look the same across several clips, make a still first and use image to video.
FAQ
- Text to video or image to video?
- Text to video is faster to start. Image to video gives you more control, because you choose the first frame. For a specific character or product, make the still first and animate it.
- Do the videos have sound?
- Some models generate audio with the video, others are silent. For speech, pair a clip with the lip sync or talking photo tools.
- How much does a video cost?
- Video is priced by the second on most models, and the studio shows the total before you run it.