Runway Gen-4: AI video with consistent characters across multiple scenes
In one sentence Runway ships Gen-4: 5-10s video generation with character, object, and environment consistency across clips. Solves the key problem for AI short-film production: the character stays itself, scene after scene.
Runway, one of the AI video pioneers, releases Gen-4. The novelty that matters isn't the quality of a single clip (already high in Gen-3 and Sora) but consistency across clips: a character you created in one scene stays recognizable in every other scene of the short.
Until now the main constraint for AI short-film makers was this: each generated clip looked like a different character, even with identical prompts. Tricks like LoRA training partially worked, but were craft-heavy. Gen-4 solves it directly: pass a "reference image" and the character shows up consistent across all scenes.
For indie filmmakers, advertising, and content creators it's a workflow shift: the "write screenplay → generate scene-by-scene → edit" pipeline becomes feasible without major hits to visual identity.
Companies
Runway
Tools
Runway Gen-4, Gen-4 Turbo
Tags
Sources