Luma AI Video Generator

Luma is Luma Labs’ video platform. Ray3.2 is the current generate model: text-to-video or image-to-video, 5 or 10 seconds, 360p to 1080p, default 720p. Modify Video edits a clip you already have. The limits below are from Luma’s Ray3.2 page.

01 — Capabilities

A 10-second generate, or an edit of a clip you already have

Ray3.2 creates. Modify Video and Reframe rewrite. They are different modes. Text-to-video and image-to-video do not make a soundtrack.

01
5 or 10 seconds on generate
Luma’s Ray3.2 page lists 5 or 10 seconds for text-to-video and image-to-video. That is the native generation ceiling. It is not a 30-second model.
02
360p to 1080p, 720p by default
You can generate in 360p, 540p, 720p (default) and 1080p. Ray3.2 also lists native 16-bit HDR and EXR export. Those are delivery formats for grading, not a higher generate resolution.
03
Modify Video up to 20 seconds
Modify Video is video-to-video. Luma lists up to 20 seconds depending on frame rate. Reframe takes a source up to 16 seconds. Both keep the original soundtrack. That is an edit, not a blank generation.
04
No native audio on a blank generate
Text-to-video and image-to-video do not produce voice, music or effects. Luma says you add those later. Only Modify Video and Reframe preserve source audio.

02 — Prompt to output

Prompts that match the clips above

Each prompt describes the picture on its card. “Use this prompt” sends that text to Create. The generator on this page runs Seedance 2.5, not Luma.

01 · Portrait
PromptClose-up of a young man with curly hair in a grey knit sweater, soft window light. He glances away, then smiles. Slow push-in, shallow depth of field.
02 · Product
PromptA white leather sneaker on a pale grey studio backdrop, soft diffused light. The camera orbits slowly. Stitching stays sharp, the look stays minimal.
03 · Street
PromptA slow dolly down a cobblestone lane in an old European village. Climbing roses on stone walls, café tables in morning sun, a few people walking in the distance.
04 · Character
PromptA young woman with shoulder-length brown hair in an olive t-shirt, late-afternoon gold light. She turns her head and looks off. Breeze in her hair, a handheld feel.

03 — Where it fits

Jobs that fit a 10-second generate or a 20-second edit

Use Ray3.2 when you start from a prompt or a still. Use Modify Video when the motion already exists and you need the picture rewritten.

A short shot with a named camera
Five or ten seconds is enough for a push-in, an orbit or a look-off. Longer sequential action will not fit the generate window.
Animate a still, keep its crop
Image-to-video takes a first-frame still. The 5 or 10 second choice is the length, not a chain of extensions.
Rewrite a clip you already shot
Modify Video keeps the source motion and changes how it looks, up to 20 seconds. Reframe is for a source up to 16 seconds. Neither starts from text alone.
Hand HDR to a colorist
16-bit HDR and EXR are listed for grading and VFX. The generate size is still 360p–1080p. Add voice and music after, because the blank generate has none.

04 — Verdict

What Luma’s Ray3.2 page actually lists

Strengths

Text-to-video and image-to-video, 5 or 10 seconds
360p, 540p, 720p and 1080p, default 720p
Modify Video up to 20 seconds; Reframe source up to 16 seconds
16-bit HDR and EXR listed for delivery

Trade-offs

Text-to-video and image-to-video have no native audio
10 seconds is the generate ceiling
Ray2 is the older, shorter line — do not read its specs onto Ray3.2
Luma is not available on ImgToVid

05 — Side by side

Luma Ray3.2, Veo 3.1 and Kling 4.0

Clip length and whether audio is generated with the picture are where these three stop lining up. Modify Video’s 20-second source is an edit cap, not a Ray3.2 duration.

	Luma Ray3.2	Veo 3.1	Kling 4.0
Developer	Luma Labs	Google DeepMind	Kuaishou
Single clip	5 or 10 seconds	4, 6 or 8 seconds	3–30 seconds
Picture	360p–1080p, default 720p	720p, 1080p, 4K at 24 fps	720p, 1080p and 4K listed
Sound	None on generate; kept on edit	Native, with the picture	Stereo, with lip sync
Edit path	Modify Video up to 20s; Reframe 16s	Not the Veo 3.1 API	Omni reference, not a rewrite of a full clip
On ImgToVid	Coming soon	Coming soon	Coming soon

06 — How it works

Start a clip on ImgToVid

Step 1
Describe one shot
Subject, action, camera. If you wanted Luma’s 5 or 10 second window or Modify Video’s rewrite, write the move in the prompt. Luma will not be the model that runs it.
Step 2
Use the generator on this page
It runs Seedance 2.5, because Luma is not offered here. The prompt and the settings carry into Create.
Step 3
Generate in Create
Membership and credits are checked there. Coming back to this page does not switch the model over to Luma.

07 — FAQ

Luma, in plain terms

What is Luma?

Luma is Luma Labs’ video platform, often called Dream Machine. The current generate model is Ray3.2 (text-to-video and image-to-video). Modify Video and Reframe edit existing clips. Ray2 is the older generate line.

How long is a Ray3.2 clip?

5 or 10 seconds for text-to-video and image-to-video. Modify Video can go up to 20 seconds depending on frame rate. Reframe takes a source up to 16 seconds.

Does Luma generate audio?

Not on text-to-video or image-to-video. Luma’s Ray3.2 page says those modes have no native audio. Modify Video and Reframe keep the source soundtrack.

What resolution does Ray3.2 generate?

360p, 540p, 720p (default) and 1080p. 16-bit HDR and EXR are listed as delivery formats, not a higher generate resolution.

Can I generate Luma videos on ImgToVid?

No. The generator on this page runs Seedance 2.5 so you can draft a shot. Luma itself is marked coming soon.

Can I use the results commercially?

Commercial use of anything you generate on ImgToVid depends on your plan. Check Pricing and our Terms before you publish, and make sure prompts and uploads do not infringe someone else’s rights.