Runway introduced Aleph in July 2025 as a model that uses an existing video as the material for further generation and editing. Its announcement describes changing objects within footage, altering lighting or style, and generating different views of a scene. That makes the input clip a starting point for visual changes, giving creators a way to develop variations of footage they already have. These are the capabilities Runway announced; the accompanying text does not quantify how consistently particular edits succeed.
Terms in this piece · Glossary
context window — The maximum amount of text a model can consider at once — its working memory for the current conversation or task.
