Ray3.2 video generation
Create text-to-video and image-to-video clips, or transform existing footage with video-to-video workflows. Ray3.2 supports output up to 1080p, with V2V workflows described for clips up to 20 seconds.
Luma APIs
Freemium
Luma APIs provide image and video generation models for creative products, workflows, and production pipelines. They expose Uni-1.1 for image generation and Ray3.2 for controllable video generation, including editing and reframing workflows.
Luma APIsとは？
Luma APIs are developer-facing image and video generation services for adding generative media to creative products, workflows, and production pipelines. The offering centers on Uni-1.1 for image generation and Ray3.2 for controllable video generation, editing, and reframing.
Ray3.2 is designed for production-oriented video workflows, with text-to-video, image-to-video, video-to-video, and reframe tasks. It supports frame-level direction through Multi-Keyframe, source-video transformations, motion and performance transfer, 1080p output, native HDR generation, and 16-bit EXR export.
Uni-1.1 is a multimodal reasoning model for generating and editing images. The API page highlights reference-guided generation, spatial reasoning, scene completion, and support for chained or messy prompts. Together, Uni-1.1 and Ray3.2 can be used in multimodal workflows that combine image creation with video generation.
The API pricing model includes pay-per-use Build access with no minimum commitment and a Scale option with dedicated capacity and an SLA. Build access is rate-limited and does not include a latency SLA. The supplied product information does not specify authentication, SDKs, integrations, or deployment procedures.
Luma APIsでできること
Multi-Keyframe direction
Set up to 16 keyframes within a single clip and specify what changes or remains consistent across the sequence, providing more control than a single start or end image.
Video transformation and reframe
Use motion transfer, camera motion transfer, character transformation, environment changes, relighting, and product swaps to adapt footage. Reframe is intended to produce variants for different aspect ratios.
HDR and EXR output
Ray3.2 offers native HDR generation and 16-bit EXR export. The source describes EXR output as suitable for color grading, compositing, and VFX workflows, including ACES2065-1 (AP0).
Uni-1.1 image reasoning
Generate or edit images with reference-guided controls, spatial reasoning, scene completion, and plausibility-driven transformations. The API page describes separate reasoning and generation endpoints.
Pay-per-use and scale options
Build access uses usage-based billing with no minimum commitment. A Scale option adds dedicated capacity, guaranteed throughput and latency, and an SLA for production workloads.
利用シーン
“Creative application development”
Embed image and video generation into products that need programmatic creation, editing, or media variation rather than relying only on an interactive interface.
“Campaign and market variants”
Generate multiple versions of a campaign from one underlying brand system. Ray3.2 features such as reframe, product swap, and environment change are suited to adapting creative across markets and formats.
“Post-production and VFX workflows”
Use video transformation, HDR generation, and EXR export when generated material must move into color grading, compositing, or VFX pipelines.
“Multimodal creative pipelines”
Combine Uni-1.1 image generation with Ray3.2 video generation to develop still references, visual concepts, and moving-image outputs within one API-based workflow.
よくある質問
Which models are available through Luma APIs?
The API page highlights Uni-1.1 for image generation and Ray3.2 for video generation. Luma describes them as complementary models for multimodal creative workflows.
What can Ray3.2 generate or transform?
The documented tasks include text-to-video, image-to-video, video-to-video, and reframing. Ray3.2 also supports keyframe direction, motion transfer, camera motion transfer, character and environment changes, relighting, and product swaps.
What video output options does Ray3.2 support?
The source describes 1080p output, SDR and HDR generation, and HDR plus 16-bit EXR export. HDR is priced at twice SDR, while HDR plus EXR is priced at three times SDR. V2V workflows are described for up to 20 seconds.
How is API usage billed?
The Build option uses pay-per-use billing with no minimum commitment. Ray3.2 pricing depends on task, resolution, duration, and output format. Build access is rate-limited and does not include a latency SLA; Scale is described as offering dedicated capacity and an SLA.
Does the supplied information explain API setup or integrations?
No. The available product information references API access and documentation but does not establish authentication steps, SDK availability, integrations, or deployment options.
クイック情報
- Category
- Developer Tool; AI Image Generation; AI Video Generation
- Primary models
- Uni-1.1 for images and Ray3.2 for video
- Video capabilities
- Text-to-video, image-to-video, video-to-video, and reframe
- Video output
- Up to 1080p; SDR, HDR, and HDR plus 16-bit EXR options
- Billing
- Pay per use for Build; Scale option with dedicated capacity and SLA
- Source domain
- lumalabs.ai
Luma APIsの代替品
SEEAPI
seeapi.com
SEEAPI is a unified platform for creating and integrating image, video, and other multimodal AI workflows across leading models. It supports browser-based experimentation and a public API for developers building creative tools, automations, and content systems.
ChatGPT Images 2.5
openai.com
ChatGPT Images 2.5 is an image-generation and editing model for turning ideas, sketches, prompts, and reference photos into more personalized images. It supports iterative creative work in ChatGPT and production image workflows through the API.
Kling AI
kling.ai
Kling AI is a multimodal creative studio for generating images and videos from text, images, and reference inputs. It also provides APIs for developers and enterprises that need image and video generation in products or workflows.
RunWeave
runweave.app
RunWeave is a unified inference API for developers building with image, video, audio, 3D, and related AI models. It provides access to 1,000+ models through one REST endpoint and one API key.
ModelsLab
modelslab.com
ModelsLab is a developer platform that provides APIs for image, video, audio, 3D, and LLM generation through a unified service. It supports teams building generative media features without integrating each model provider separately.
Apiframe
apiframe.ai
Apiframe is a unified REST API for generating AI images, videos, and music. It helps developers and automation teams add media generation through one API, with asynchronous jobs, webhooks, SDKs, and CDN-hosted outputs.