Luma Ray 3.2
videoLuma's Ray video model on Visxy. A clip from a prompt, or from an opening frame and an optional closing frame. No sound switch.
What it supports
- Aspect ratios
- 3:4, 4:3, 1:1, 9:16, 16:9, 21:9
- Resolutions
- 540p, 720p, 1080p
- Duration
- 5–10s
What it costs
- 540p
- 15 credits per second
- 720p
- 30 credits per second
- 1080p
- 60 credits per second
Credits come with a Generator plan. The exact cost is shown on the Generate button before anything is spent.
Luma Ray 3.2 is Luma's Ray video model. It is not Ray 2. Use it for a directed clip from a prompt or from frames you already have.
What it is good at
Luma describes Ray3 as a video model built with people in film, advertising, and games: it reasons about the prompt, follows instructions, and aims at physically plausible motion. On Visxy you direct it with the prompt and the frames.
How to use it on Visxy
Write the shot. Give one picture as the opening frame and an optional second as the closing frame. There is no sound switch.
Where it differs
Kling 3.0 has the same opening-and-closing shape and adds a sound switch. MiniMax H3 matches the frames and adds a reference-picture mode. Ray is the Luma model, and it stays with the prompt and the two frames.
FAQ
Can I set a first and last frame?
Yes. The second picture is optional and closes the clip.
Is there sound?
No sound switch on this model.
Is this Ray 2?
No. Luma's model on this page is Ray 3.2.
Can I attach several pictures?
No. One picture opens the clip, and a second can close it.
Official sources
More video models
All video modelsKling 2.6 Motion Control
Kuaishou's Kling 2.6 motion transfer. One picture of a subject plus one clip whose motion it copies. A prompt is optional.
Kling 3.0 Motion Control
Kuaishou's Kling 3.0 motion transfer. One picture of a subject plus one clip whose motion it copies. A prompt is optional.
Wan 3.0
Alibaba's Wan 3.0 video model. A longer clip from a prompt or two frames, with optional sound, or from reference pictures you name.
Wan 3.0 Prime
Alibaba's faster Wan 3.0. The same prompt, frames, sound switch, and reference pictures, aimed at a quicker clip.
Vidu Q4
Shengshu's Vidu Q4. A clip that starts from a picture, with optional sound, or from many reference pictures you name in the prompt.
Seedance 2.0 Mini
ByteDance's smaller Seedance model. A clip from a prompt or from an opening frame and an optional closing frame, with sound that starts off.