Ray 3.2 AI Video Generator

What It Does

Ray 3.2 is Luma AI's latest AI video generation model that transforms text descriptions, images, and existing footage into fluid, cinematic video clips. Unlike traditional AI video tools, Ray 3.2 offers frame-level control with up to 16 keyframes per clip, native 1080p HDR output, and professional 16-bit EXR export in ACES2065-1 color space—making it production-ready for VFX pipelines, advertising campaigns, and professional post-production.

Who It's For

Ray 3.2 is built for filmmakers, advertising teams, VFX artists, game studios, content creators, and marketing agencies who need professional-quality video without expensive production shoots. It serves anyone who wants to prototype concepts quickly, create campaign variants at scale, or generate platform-specific content from existing footage.

Core Workflows & Features

Ray 3.2 provides four distinct video generation workflows:

• Text to Video: Generate cinematic clips from detailed scene descriptions with accurate motion, camera behavior, and lighting
• Image to Video: Animate still images into motion while preserving visual identity—ideal for concept art and character shots
• Video Edit (Modify Video V2): Transform existing footage while preserving original motion, timing, and scene structure—swap environments, change outfits, restyle lighting while keeping performances intact
• Video Reframing: Change aspect ratios (9:16, 16:9, 1:1, 4:3, 21:9) without cropping, using scene-aware fill that matches mood and lighting

Production-Grade Capabilities

Ray 3.2 handles demanding visual production tasks including multi-keyframe control across clips, expressive facial performance tracking (up to 8 faces simultaneously), motion transfer between scenes, character transformation, environment changes, relighting, product swaps for SKU variants, and campaign localization for market-specific video content.

Technical Specifications

• Native 1080p resolution across all workflows (no upscaling)
• True 16-bit HDR color generated by the model (not post-processed)
• 16-bit EXR export ready for Nuke, DaVinci Resolve, and VFX pipelines
• Up to 20-second clips for Video Edit mode
• 5s or 10s generation for T2V and I2V workflows
• Up to 16 keyframes per clip for precise directorial control
• Full API access for pipeline integration

Use Cases & Industries

Ray 3.2 serves film previsualization for shot prototyping, advertising and brand video for product variants and commercial concepts, game and animation for concept prototyping and cutscenes, social media content creation with vertical format support, VFX and post-production with professional color workflows, and agency pipelines requiring automated video generation at scale.

What Sets It Apart

Unlike older AI video models, Ray 3.2 prioritizes controllable video transformation over simple text-to-video generation. It's designed for production environments where existing footage needs to be remixed, restyled, or repurposed—reducing reshoots and enabling rapid iteration. The combination of multi-keyframe control, native HDR, EXR export, and API access makes it suitable for professional pipelines rather than just experimentation.