One click, whole workflow
Recipes are here. Package a whole workflow and run it in one click
Explore recipesRunway Video
- Gen-4
- Gen-4.5
byRunway
Director-grade camera choreography and seamless frame interpolation for cinematic motion
How Runway Video works
Direct continuous 5 to 10-second cinematic sequences by combining descriptive camera trajectories with optional start and end image prompts.
Anchor start and end frames
Upload an initial image baseline or pair first and last keyframes to govern character likeness and lighting continuity across the take.
Direct camera and motion cues
Describe specific camera trajectories like slow push-ins or tracking arcs alongside subject dynamics and progressive environmental changes.
Generate and export takes
Select 5-second or 10-second durations in cinematic widescreen or vertical formats, rendering temporal consistency with physical inertia.
What Runway Video is good at
From rapid storyboarding on Turbo to flagship Gen-4.5 frame interpolation, explore dedicated tools for director-level temporal control.
Made with Runway Video
A gallery of high-fidelity camera choreography, physical momentum, and deliberate lighting continuity generated across landscape and portrait ratios.
What people build with Runway Video
Filmmakers, creative directors, and motion designers leverage fluid frame generation to build commercial spots, previsualization takes, and visual campaigns.
Runway Video specs
Plans and credits- Versions
- Gen-4 Turbo, Gen-4.5
- Inputs
- Prompt (text), Image Prompt (image), End Image Prompt (image)
- Output
- video
- Duration
- 5 seconds, 10 seconds
- Aspect ratios
- 1280:720 Horizontal, 720:1280 Vertical, 1104:832 Landscape, 832:1104 Vertical, 960:960 Square, 1584:672 Landscape
- Cost
- 345–1,653 credits per run
Frequently asked questions
Gen-4 and Gen-4.5 are the high-tier video synthesis architectures powering Runway Video. They specialize in temporal consistency, cinematographic camera control, and keyframe-conditioned animation, enabling creators to generate 5-second and 10-second cinematic takes from structured text prompts or reference images.
Gen-4 Turbo is optimized for rapid generation speed and lower compute cost, making it ideal for animatics, storyboarding, and fast composition tests. In contrast, Gen-4.5 is the flagship tier built for maximum temporal stability, intricate first-to-last frame morphing, and meticulous physical consistency in production workflows.
Use text-to-video when generating entirely new scenes from scratch without pre-existing concept art or photographic assets. When you require strict character continuity, brand styling, or an established lighting palette, connect an initial image prompt to anchor the generation before directing camera motion.
Dual-image conditioning smoothly interpolates motion from a starting keyframe to an end destination keyframe over a 10-second duration. To maintain seamless visual coherence, both frames should share compatible lighting angles, camera perspective, and subject framing.
The model excels at controlled cinematic camera maneuvers—such as slow push-ins, tracking dollies, orbital arcs, and panoramic sweeps—paired with natural environmental dynamics like drifting mist, flowing fabric, and realistic physical momentum.
Try Runway Video on Fuser
Director-grade camera choreography and seamless frame interpolation for cinematic motion