Luma AIVideo model
Ray 3.2
10slongest single cliphigher is better
Summary
Ray 3.2 is a video model from Luma AI. Ray 3.2 supports text-to-video and image-to-video; the API also documents video editing, reframing, and extension.
On this page
Video sample from our test
This video exists as a documented run for the exact model version. Playback is muted.
Motion and physical consistency
This sample tests Anatomy, object consistency, camera tracking, rain, and reflections.
Prompt and settings
Generate exactly one continuous five-second 16:9 cinematic video. A fictional bicycle courier in a mustard-yellow raincoat rides a red city bicycle from left to right across a rain-wet Hamburg street at blue hour for the entire clip. The camera tracks her smoothly at waist height. Both hands remain visible on the handlebars, the bicycle geometry stays physically correct, and a transparent umbrella fixed to the bicycle moves naturally in the wind. Wet-street reflections and raindrops follow realistic physics. No cuts, no slow motion, no text, no logos, no subtitles, and no audio.
Generated through fal.ai with model luma/agent/ray/v3.2/text-to-video. Archived settings were 16:9, one request, first output, prompt motion-and-physics, requested 5 s, provider-default audio (no switch), seed provider random, request {"prompt":"Generate exactly one continuous five-second 16:9 cinematic video. A fictional bicycle courier in a mustard-yellow raincoat rides a red city bicycle from left to right across a rain-wet Hamburg street at blue hour for the entire clip. The camera tracks her smoothly at waist height. Both hands remain visible on the handlebars, the bicycle geometry stays physically correct, and a transparent umbrella fixed to the bicycle moves naturally in the wind. Wet-street reflections and raindrops follow realistic physics. No cuts, no slow motion, no text, no logos, no subtitles, and no audio.","duration":"5s","resolution":"720p","aspect_ratio":"16:9"}, configured price estimate $1.00, enforced estimate $1.00.
Head-to-head comparisons
Show all comparisons (94)
All specifications
Input and output
Licence and access
API
Further details
- Input modes
- Text-to-video and image-to-video; the API also documents video editing, reframing, and extension.
- Source: Input modes
- Generation limits
- For documented generation modes: 540p, 720p, or 1080p and 5 or 10 seconds. The launch announcement mentions up to 20 seconds, but the current model guide does not state this as a general T2V limit.
- Source: Generation limits
- Audio output
- Native audio output is not clearly documented in the current model materials reviewed.
- Source: Audio output
Prices
- fal.ai (luma/agent/ray/v3.2/text-to-video)
- $1 per request (estimated)Source
No Luma AI subscription includes this model. Through other providers from ≈ €21.30 a month (Recraft Pro, with limited credits). 7 plans in Germany.
| Plan | Monthly price | Note |
|---|---|---|
| Recraft Pro | ≈ €21.30 | with limited credit |
| Recraft Teams | ≈ €23.43 | with limited credit, for teams, per seat |
| Luma Plus | ≈ €31.95 | with limited credit |
| Luma Pro | ≈ €95.86 | with limited credit |
| Luma Ultra | ≈ €319.55 | with limited credit |
| Luma Enterprise | n/a | for enterprises |
| Luma Team | n/a | for teams |
Frequently asked questions about Ray 3.2
How long can a video with Ray 3.2 be?
Up to 10 seconds in one go. Resolution: 1080p.
According to Luma AI
- Longest clip
- 10s
- Resolution
- 1080p
- Open weights
- no
Where can I use Ray 3.2?
Not in a Luma AI subscription. Including other providers that makes 7 plans in Germany.
Subscription comparison, prices in Germany incl. VAT
At Luma AI
No Luma AI subscription includes this model.
Through other providers
7 plans from 2 providers, from ≈ €21.30 a month (Recraft Pro), with limited credits.
Sources and data date
| Type | Evidence and data date |
|---|---|
| Additional source | Luma AI (retrieved November 1, 2026) · Luma AI · Editorial description reviewed November 1, 2026 |
| Additional source | Luma AI (retrieved November 1, 2026) |
| Additional source | fal.ai |