Luma AI model

Luma Ray 2

Name: Luma Ray 2
Brand: Luma AI
Availability: InStock

Use Ray 2 when the shot is moving past draft territory and you want the premium Luma tier for new cinematic shots, controlled still animation, source-video modifications, and final delivery reframes without splitting the workflow across separate model pages.

Best for premium cinematic work when you want generation and source-video edits inside one Luma surface.

Text->VideoImage->VideoModify from source clipReframe to 1:1 or 9:16Generate: 5s or 9s540p to 1080p on generate

Start generating with Luma Ray 2 Compare Ray 2 vs Ray 2 Flash ->

Pay-as-you-go · Price shown before you generate

Luma Ray 2 AI video example: **Scene (plain language):** A glass bottle of sparkling citrus drink sits on a sunlit outdoor café table in... — Price$0.13/s
Duration5s
Format16:9
View render →

Best use cases

Premium cinematic concept filmsFashion and product hero motionStill-to-video with guided endingsSource-video look changes without new blockingLandscape-to-vertical social repackagingLoop-ready luxury and ambient shots

Why Luma Ray 2 is strategically useful

One model, real workflow coverage (Generate new clips, animate stills, modify existing footage, and reframe approved shots from the same Luma page.)
Modify is not hidden anymore (The Fal adherence, flex, and reimagine presets are exposed directly so teams can steer how far the edit should move from the source clip.)
Reframe is practical, not cosmetic (Square and vertical outputs matter when the approved master has to ship across paid social, mobile, and commerce placements.)
Cleaner family story (Ray 2 stays the premium tier, Ray 2 Flash stays the fast tier, but both models now expose the real public workflow surface.)

Real specs - Luma Ray 2 in MaxVideoAI

The limits that shape your renders.

Price / second540p $0.13/s720p $0.26/s1080p $0.52/s

Text-to-Video

Image-to-Video

Video-to-VideoSupported (modify / reframe workflows)

Reference image / style referenceSupported (single start image)

Reference videoSupported (source clip for modify / reframe)

Max resolution1080p

Max duration9s max

Aspect ratios16:9 / 9:16 / 1:1 / 4:3 / 3:4 / 21:9 / 9:21

FPS options24

Output formatjpg / jpeg / png / webp / mp4 / mov / webm / m4v / gif

WatermarkNo (MaxVideoAI)

Generate workflowsDetails

Text-to-Video and Image-to-Video are the creation modes: 5 or 9 seconds, 540p to 1080p, multiple cinematic aspect ratios, optional loop, and optional end-image guidance for image-driven runs.

Use Text-to-Video for net-new cinematic beats.
Use Image-to-Video when the first frame must stay on-brand.
Add an end image only when the landing composition matters.
Use loop for ambient or product motion tests.

Modify and ReframeDetails

The workflow expansion is not just marketing copy. Ray 2 now exposes source-video editing directly, which changes how useful the model is in production.

Modify preserves timing and camera path while changing look, styling, or scene interpretation.
Adhere presets stay close to the source clip, while flex and reimagine allow more creative drift.
Reframe keeps the motion but changes delivery format, with square and vertical outputs supported.
Optional crop and grid controls help keep the priority subject inside the new frame.

Luma Ray 2 examples

Use the examples rail to compare net-new generation, still animation, modify passes, and delivery reframes inside the same Ray 2 family.

Luma Ray 2 · 5s

Recreate this shot ->

Luma Ray 2 AI video example: **Scene (plain language):** A young man in bold streetwear runs through a crowded 1990s-style night market u...

Luma Ray 2 · 5s

Recreate this shot ->

Luma Ray 2 AI video example: Vertical 9:16 TikTok-style UGC selfie video, handheld smartphone feel, natural indoor daylight near a window...

Luma Ray 2 · 5s

Recreate this shot ->

Luma Ray 2 AI video example: Wide 16:9 full-body unboxing video in a clean studio/kitchen setting. A person is fully visible (head-to-toe...

Luma Ray 2 · 5s

Recreate this shot ->

Luma Ray 2 AI video example: Wide 16:9 cinematic action shot, a runner sprints through a rainy city street at night, water splashes reali...

Luma Ray 2 · 5s

Recreate this shot ->

View all Luma examples ->

How to Write a Great Video Prompt

Developers guide

Works best when you brief it like a cinematographer: one clear shot, simple timing, and visible actions.

Tip: duration + aspect ratio are set in the UI — your prompt controls subject, action, camera, lighting, style, and the visual ending.

Quick prompt (fast iteration)

Use 1–2 sentences when you want variations.

Quick = variations. Use for fast iteration.

Template (copy/paste)

[Style] + [Subject doing 1 clear action] + [Where] + [Camera move] + [Lighting] + [Ending / visual payoff]

Example

Handheld smartphone UGC clip of a woman unboxing a new skincare bottle at a kitchen table. She peels the seal, smiles, and turns the bottle toward camera. Soft window daylight, natural colors, clean kitchen background, product label held readable in the final beat.

Structured prompt (best for reliable results)

Separate information so the model can follow it consistently.

Structured = consistency. Use when you need reliable results.

Template (copy/paste)

Scene (plain language):
[Subject + setting + props + time of day. Add 2–3 distinctive visual anchors.]

Cinematography:
- Camera shot: [wide / medium / close-up, angle]
- Camera motion: [slow push-in / handheld / pan / tracking]
- Lens look + depth of field: [e.g., 35mm, shallow DOF]
- Lighting + palette: [key light + 3 palette anchors]

Actions (beats):
- [Beat 1: a small, visible action]
- [Beat 2: another clear beat]
- [Beat 3: a final beat in the last second]

Constraints:
No logos, no readable text, no subtitles/overlays.

Pro prompt (ultra-specific "film crew brief")

Use this when you need a very specific cinematic look or continuity across shots.

Pro = continuity. Use for precise, repeatable looks.

Template (copy/paste)

Project / intent:
[One-line goal. What should the viewer feel/understand?]

Subject:
[Who/what. Wardrobe/materials. 2-3 distinctive traits.]

Location / set:
[Where + time of day + weather. Add 3 visual anchors (specific nouns).]

Cinematography:
- Framing: [wide / medium / close-up] + [angle]
- Lens feel + depth of field: [e.g., 35mm natural, shallow DOF]
- Camera movement: [ONE move: slow dolly-in / handheld / pan / tracking]
- Composition: [centered / rule of thirds / negative space]
- Look (optional): [clean digital / subtle film grain / soft bloom]

Lighting & color grade:
- Key light: [soft window / golden hour / neon practicals / studio key]
- Contrast: [low / medium / high]
- Palette anchors: [3-5 anchors: "warm sunrise, teal shadows, amber highlights"]

Action (timed beats):
- Beat 1 (start): [visible action + camera behavior]
- Beat 2 (middle): [visible action + camera behavior]
- Beat 3 (end): [final action + end pose / reveal]

Constraints:
No logos. No readable text. No subtitles/overlays. No slow-motion. No jump cuts.

Storyboard prompt (multi-shot / shot list)

Use this when you want a mini-story in one clip. A storyboard prompt (aka multi-shot / shot list prompt) gives clear timing, camera direction, and continuity. Also called shot-list or sequenced prompts.

Storyboard = beat timing. Use for mini-stories in one clip.

Template (copy/paste)

Storyboard / shot list prompt
Duration: [4/8/12s] • Aspect: [16:9 or 9:16]

Scene + continuity:
[Same subject + same location + same wardrobe/props + same lighting throughout.]

Shot 1 (0–2s):
[Framing + subject action + camera move]

Shot 2 (2–6s):
[Framing + subject action + camera move]

Shot 3 (6–8/12s):
[Framing + final action/reveal + camera move or settle]

Lighting + mood:
[Golden hour / soft daylight / neon night… + 2–3 palette anchors]

Constraints:
No logos. No readable text. No subtitles/overlays. No jump cuts. No slow-motion.

Practical strengths and boundaries

What works best

High-value Luma workflow surface: generate, modify, and reframe in one place.
Useful when premium cinematic quality matters but the delivery plan also requires source-video edits.
Better family clarity than exposing separate public modify or reframe pseudo-models.

Common problems → fast fixes

Feels random / inconsistent → simplify to: subject + action + camera + lighting. Re-run 2–3 takes.
Motion looks weird → reduce movement: one camera move, slower action, fewer props.
Subject drifts off-brand → start from a reference image and lock palette + lighting.
Text looks wrong → avoid readable signage, tiny UI, micro labels. Keep text off-screen.

Hard limits to keep in mind

5s and 9s plus 540p to 1080p apply to the generate modes, not to source-video edit durations.
Modify and Reframe still depend on the quality and clarity of the uploaded source clip.
Ray 2 is the premium tier, so Flash remains the better option for fast early-stage iteration.

Compare Luma Ray 2 vs other AI video models

Not sure if Luma Ray 2 is the best fit for your shot? These side-by-side comparisons break down the tradeoffs — price per second, resolution, speed, and motion style — so you can pick the right engine fast.

Each page includes real outputs and practical best-use cases.

Luma Ray 2 vs Google Veo 3.1 Fast

Use Veo 3.1 Fast for affordable text prompts, start-image animation, optional last-frame control, and extend workflows with optional native audio inside one unified MaxVideoAI model page.

Compare Luma Ray 2 vs Google Veo 3.1 Fast →

Luma Ray 2 vs Kling 3 Standard

Use Kling 3 Standard for multi-prompt sequences, element references, and voice controls at a lower $/s rate in 1080p.

Compare Luma Ray 2 vs Kling 3 Standard →

Luma Ray 2 vs Seedance 1.5 Pro

Generate Seedance 1.5 Pro clips with cinematic motion, camera lock, and native audio. Supports text-to-video or image-to-video up to 12s.

Compare Luma Ray 2 vs Seedance 1.5 Pro →

Safety & people / likeness

Don’t generate real people or public figures (celebrities, politicians, etc.).
No minors, sexual content, hateful content, or graphic violence.
Don’t use someone’s likeness without consent.
Some prompts and reference images may be blocked — generic characters and scenes are fine.

FAQ

What workflows does Luma Ray 2 expose in MaxVideoAI?

Ray 2 keeps the full public Fal workflow surface inside one model page: text-to-video, image-to-video, modify, and reframe.

Which options are exposed in Modify?

Modify uses a source video plus an optional prompt and optional reference still. MaxVideoAI exposes the Fal strength presets from adhere_1 through reimagine_3.

Which options are exposed in Reframe?

Reframe uses a source video, a target aspect ratio, an optional prompt, an optional reference still, plus optional grid and crop-window controls.

When should I choose Ray 2 over Ray 2 Flash?

Choose Ray 2 when you want the flagship Luma tier for higher-confidence finals, while Ray 2 Flash is better for faster drafts and cheaper iteration.