
Ranked #1 on Artificial Analysis. Runway's flagship cinematic model.
Runway Gen-4.5 is the latest from one of the original pioneers of AI video. Delivers the highest coherence of any model in the Runway family, with cinematic colour science, smooth temporal consistency, and strong prompt adherence. Supports text-to-video and image-to-video at up to 1080p.

“A slow camera walkthrough of a brutalist concrete museum, morning light raking across board-marked walls, ambient footstep audio”

“A matte black wireless speaker rotating slowly on a marble plinth, dramatic side lighting, clean studio backdrop, product reveal pacing”

“A silver electric SUV driving along a coastal cliffside road at golden hour, tracking shot from a drone, smooth motion, warm lens flare”

“A model in a structured linen coat walking through a sun-dappled olive grove, editorial slow motion, warm Mediterranean palette”

“A camera pushing through a double-height living room with floor-to-ceiling glazing, late afternoon light casting long shadows across oak flooring”

“A cinematic game trailer opening: a lone figure stands on the edge of a floating island, clouds swirling below, orchestral swell building”
Type a detailed prompt describing the video you want, or upload a reference image as a starting frame.
Pick your resolution and duration. See the credit cost before you generate.
Your video is ready in 1-3 minutes. Download, iterate, or extend the sequence.
Jump into the Studio and start generating. Plans from £10/month.
Choose a PlanRunway were among the first to put AI video generation into the hands of creative professionals, and Gen-4.5 is their most capable model to date. It holds the #1 position on the Artificial Analysis Text to Video benchmark with 1,247 Elo, ahead of every other model on the market. It delivers the highest coherence in the Runway family: characters hold their identity across frames, camera moves track smoothly, and complex multi-object scenes stay internally consistent. For design professionals who need footage that holds up under scrutiny, coherence is the quality that separates usable output from throwaway drafts.
The colour science in Gen-4.5 leans cinematic. Skin tones render naturally, lighting transitions feel graded rather than computed, and shadow detail holds without banding or crush. Temporal consistency is strong across the full clip duration. Objects maintain their shape, texture, and spatial position frame to frame. This matters most in product films, architectural walkthroughs, and brand content where a single wobble or colour shift breaks the illusion.
For client-facing deliverables, Gen-4.5 closes the gap between AI-generated footage and professionally shot video. Product reveal films, brand content pieces, architectural fly-throughs, and editorial fashion clips all benefit from the model's combination of coherence, colour accuracy, and prompt fidelity. Generate at 1080p for presentation-ready output, or draft at 720p and promote the best takes to higher resolution.
Gen-4.5 tracks objects, characters, and environments with a level of consistency that previous Runway models could not match. Characters keep their features. Products hold their form. Environments stay spatially accurate as the camera moves. This is the quality that determines whether AI video is usable in professional workflows or limited to social media experiments.
The model produces footage with natural colour grading, smooth highlight roll-off, and shadow detail that holds without noise or banding. Skin tones, metallic surfaces, fabric textures, and natural materials all render with the tonal accuracy that design professionals expect from graded footage.
A small indie studio building creative tools the way they should be built. No VC theatre, no funnel games, no faceless support.
Professional video generation. Plans from £10/month.