MODELS
Ray2
by Luma AI
Overview
Large-scale generative video model for realistic visuals, coherent motion, and text/image-to-video workflows through Luma’s API.
Details
Ray2 is Luma AI’s large-scale generative video model. Luma announced Ray2 on January 15, 2025 with text-instruction understanding, coherent motion, higher-resolution support, 5–10 second clips, extensions, keyframes, looping, and API access for developers and enterprise users. Official Luma pages describe it as supporting realistic visuals and text/image-to-video workflows through Luma’s API.
When to Use
Generate short realistic video clips from text or image inputs using Luma’s hosted API. Use when a workflow needs coherent motion keyframes looping clip extension or higher-resolution video generation. Evaluate for developer or enterprise video-generation integrations where a commercial hosted API is acceptable.
Getting Started
- Review the Ray2 product page and Luma video-generation documentation.
- Use the Luma API documentation to understand request parameters for video generation workflows.
- Check the official changelog for Ray2 capabilities such as 5–10 second clips
- keyframes
- looping
- extensions
- and upscaling.
- Review Luma’s content policy before building production workflows.
Key Features
- •Text-instruction understanding for video generation.
- •Text-to-video and image-to-video workflows through Luma’s API.
- •Coherent motion and realistic visual generation.
- •Support for 5–10 second clips
- •extensions
- •keyframes
- •and looping.
- •Higher-resolution support
- •with API changelog coverage for Ray2 4K upscaling.
Capabilities
- •text-to-video
- •image-to-video
- •video-generation
- •video-extension
- •keyframed-video
- •looping-video
- •video-upscaling
- •hosted-api
Last updated Jun 9, 2026