Restyling existing footage
Ray3.2 is designed to change the world, look, characters, or materials in a clip while retaining useful source motion and shot structure.
Independent tool overview
Luma Labs at a glance
Luma AI is a creative workspace for generating, transforming, and finishing images, video, and audio. Its current lineup centers on Ray3.2 video-to-video transformation, UNI-1.1 image creation, agent-guided workflows, and access to selected third-party models.
Visit the official Luma Labs site ↗
- Current video model
- Ray3.2, focused on video-to-video transformation
- Current image model
- UNI-1.1
- Primary workspace
- Luma App at app.lumalabs.ai
- Ray3.2 input
- A source video, with optional prompt and keyframes
- Ray3.2 duration
- Matches the source clip, up to 20 seconds
- Starting paid plan
- $30 per month for Plus
Overview
What Luma Labs is
Luma AI has evolved beyond the older Dream Machine, Ray2, Photon, and phone-based 3D-capture framing that still appears in many descriptions. The current Luma App combines Luma's own models with agent-guided creation, editing, audio tools, project boards, and selected third-party image and video models.
The most important distinction is that Ray3.2 is a video-to-video model: it transforms existing footage while preserving useful motion, framing, and timing. It requires a source video and can use prompts or frame-specific keyframes; creators who need text-to-video or image-to-video should select another available model instead.
Use cases
Who Luma Labs is best for
The strongest fit depends on the job you need the product to complete, not the size of its feature list.
Creative campaign variants
Teams can create alternate visual treatments and localized versions from a controlled source rather than regenerating every shot from scratch.
Mixed-model creative work
The Luma App exposes Luma models alongside selected third-party models, making it useful for teams that want to compare routes inside one workspace.
Image-to-video production
Creators can develop images with UNI-1.1, organize them on Boards, and use an appropriate in-app video model to animate or extend the concept.
Fast previsualization and social content
Agent-guided generation, editing, reframing, audio, and upscale tools can shorten the path from an idea to a reviewable short-form draft.
Capabilities
Core Luma Labs features
Ray3.2 video transformation
Transforms a supplied video into a new visual treatment while using the original clip as the timing and motion foundation.
Prompt and keyframe guidance
A transformation can be guided with natural-language direction, frame-specific keyframes, or both.
Adherence controls
Controls balance preservation of source motion, structure, faces, bodies, and poses against the degree of visual reinvention.
UNI-1.1 image creation
Luma's current multimodal image model supports image generation and visual development inside the same creative workspace.
Luma Agents
A conversational interface can help plan generations, coordinate longer sequences, select tools, and iterate on creative instructions.
Third-party model access
The app also integrates selected models from other providers, expanding the available image and video generation routes.
Video editing tools
Available operations include trimming, extending, modifying, reframing, extracting frames, and upscaling supported outputs.
Audio generation
The workspace can create music, sound effects, and voiceovers through integrated audio providers.
Boards and project context
Boards organize references, outputs, and related creative work so iterations remain attached to a project.
Character, style, and visual references
Reference inputs help steer identity, aesthetic direction, composition, and visual continuity across supported workflows.
HDR-oriented Ray workflows
The wider Ray family includes HDR and higher-bit-depth output options for teams whose finishing workflow can use them.
Separate developer platform
Luma provides API access with separate keys, credits, model availability, documentation, and billing from the consumer app subscription.
Process
How the Luma Labs workflow works
Step 1
Define the deliverable
Set the shot list, aspect ratios, duration, rights requirements, approval standard, and budget before generating.
Step 2
Choose the model intentionally
Use Ray3.2 when transforming an existing clip, UNI-1.1 for images, and a suitable text-to-video or image-to-video model when no source footage exists.
Step 3
Prepare controlled inputs
Use clean source footage, licensed references, concise prompts, and exact keyframes at moments where composition or character direction matters.
Step 4
Test at low cost
Run short or economical variations first, then adjust adherence and transformation strength before committing more credits.
Step 5
Inspect the entire result
Review every frame and audio element for identity drift, continuity errors, unwanted objects, unsafe content, and rights issues.
Step 6
Finish outside the generator
Complete editorial timing, color, sound mix, captions, legal review, disclosure, and final export in the appropriate production tools.
Cost
Luma Labs pricing and free plan
Luma's individual plans are Plus, Pro, and Ultra. The app uses credits and agent-usage allowances; monthly subscription credits generally reset rather than rolling over. API keys and API billing are separate from the app subscription, and exact model costs can change.
Free evaluation
$0 with limited access
A limited way to try the app before subscribing.
- Availability and included credits can change
- Check current watermark and usage-right terms before publishing
Plus
$30/month or $300/year
Entry paid plan for individual creative use.
- Includes Luma and selected third-party image and video models
- Includes guest editing and commercial use
Pro
$90/month or $900/year
Higher-usage individual plan.
- Includes everything in Plus
- Advertised with 4x usage through Luma Agents
Ultra
$300/month or $3,000/year
Highest listed individual plan.
- Includes everything in Pro
- Advertised with 15x usage through Luma Agents
Team and Enterprise
Contact sales
Business plans for shared administration, governance, and custom needs.
- Team features include projects, organizations, sharing, usage analytics, shared credits, and SSO
- Enterprise options include commitments, training, and custom fine-tuning
Pricing checked . Check current pricing at the source ↗
Assessment
Luma Labs strengths and limitations
Where it stands out
- Ray3.2 starts from real footage, which can preserve valuable camera motion, blocking, and timing
- Prompt and keyframe controls support directed video-to-video changes
- Luma's image, video, editing, audio, and agent tools share one workspace
- Boards keep references and generations grouped by project
- Third-party integrations provide more than one generation route
- Current paid pricing is published clearly for individuals
- Commercial use is included on the current Plus plan
- Team and Enterprise tiers add collaboration and governance options
- The separate API supports programmatic creative workflows
- Legacy Genie and Capture assets remain accessible through Luma's separate 3D product surface
What to consider
- Ray3.2 requires source footage; it is not the model to choose for text-to-video or image-to-video generation
- Ray3.2 output duration follows the source and is limited to 20 seconds
- Stronger transformations can reduce preservation of camera motion, identity, structure, or fine detail
- Prompts, source footage, references, and keyframes can conflict and produce unpredictable compromises
- Generated people, hands, text, physics, lip movement, and object continuity still require frame-by-frame review
- Upscaling or HDR output does not restore genuine detail that was never generated correctly
- Integrated third-party models can have different providers, terms, data handling, costs, and availability
- Agent-selected routes may change, so production teams should record which model and settings created an approved asset
- Monthly app credits do not function like permanent stored value and may reset at renewal
- Top-up credits, subscription credits, and API credits have different rules
- The API has separate billing and may not expose the same current model catalog as the Luma App
- Plan names, model allowances, and per-generation credit costs have changed over time and should be rechecked before budgeting
- Free access should not be assumed to include the same commercial rights or watermark policy as a paid plan
- Enterprise review is appropriate when confidential inputs, private training data, retention controls, or contractual protection matter
- The workspace does not replace a full nonlinear editor, compositing suite, color system, or digital audio workstation
- Creators need rights and consent for source footage, likenesses, voices, music, brands, and reference images
- Synthetic-media disclosure, provenance records, and platform labeling may be required
- Older Dream Machine, Ray2, Photon, and Free/Lite/Unlimited documentation can be outdated relative to the live Luma App
- Luma's older Genie and Capture 3D accounts and assets are separate from the current generative workspace
Compare
Luma Labs alternatives
The right alternative depends on the specific output, workflow, controls and budget your project requires.
Content Creator
Runway Gen-4.5
Consider Runway for a broader video-production platform with a different model and editing ecosystem.
Explore Runway Gen-4.5 →Content Creator
LTX Studio
Consider LTX Studio when storyboard, shot planning, and longer narrative development are central.
Explore LTX Studio →Content Creator
Flow
Consider Flow for Google's filmmaking workflow and access to Google's generative video stack.
Explore Flow →Content Creator
Krea
Consider Krea for rapid visual exploration, enhancement, and a different multi-model creative interface.
Explore Krea →Questions
Luma Labs FAQs
What is Luma AI?
Luma AI is a creative workspace for generating and transforming images, video, and audio with Luma's own models, Luma Agents, editing tools, and selected third-party models.
Is Ray3.2 a text-to-video model?
No. Ray3.2 is designed for video-to-video transformation and requires a source clip. Choose another supported model when starting from text or a still image.
How long can Ray3.2 videos be?
Ray3.2 output follows the source clip's duration, up to 20 seconds, according to Luma's current model guide.
What happened to Dream Machine, Ray2, and Photon?
They describe earlier generations of Luma's product and models. The current Luma App centers on Luma Agents, Ray3.2 for video transformation, and UNI-1.1 for images, while older documentation may remain online.
How much does Luma AI cost?
Current individual plans are Plus at $30 per month, Pro at $90 per month, and Ultra at $300 per month, with discounted annual billing. Team and Enterprise pricing is sales-assisted.
Can I use Luma AI commercially?
The current pricing page includes commercial use with Plus and higher plans. Teams should still review the applicable terms and clear all rights in their prompts, source footage, references, voices, music, and outputs.
Are Luma App credits and API credits the same?
No. The app subscription and developer API use separate access, credits, model catalogs, and billing. Verify current API documentation before planning an integration.
Does Luma AI still support 3D capture?
Luma's Genie and Capture 3D products remain available through a separate product surface with separate accounts and assets. They are no longer the main focus of the current Luma App.
Bottom line
Our Luma Labs verdict
Luma AI is strongest when a creator wants to transform existing footage with Ray3.2 or move among image, video, audio, and agent-guided tools in one workspace. The current product is substantially broader than its older 3D-capture and Dream Machine identity. It is less suitable when a team needs deterministic frame-level control, a full post-production suite, or one consistent model and policy boundary across every generation.
Visit Luma Labs website ↗Similar Content Creator Tools
View all Content Creator AI tools →
Midjourney
Midjourney: Is a leading ai art generator that creates stunning visuals from imaginative prompts.
Gemini
Gemini: Is google’s advanced ai model for search, productivity, coding, and creative tasks.
Argil
Argil: Generate pre-edited video shorts using your AI-clone.
Flux
Flux: Is an ai design and collaboration tool that turns ideas into high-fidelity visual content.
AI training for the future of work.
Get access to all our AI courses, hundreds of real-world AI use cases, live expert-led workshops, an exclusive network of AI early adopters, and more.
AI Courses
Get unlimited access to all of our current & upcoming industry-specific AI courses for the duration of your subscription.
Daily Guides
To keep up with the rapid pace of AI, our team publishes AI implementation guides daily. Our library contains 300+ practical use cases to automate real-world work.
Workshops
Join weekly, live, interactive sessions with industry leaders who are at the forefront of AI for hands-on implementation guidance and exclusive insights.
Community
Network with an exclusive community of AI-first professionals who are working smarter with AI. Learn how early adopters are using AI in their work and businesses.