Runway launches Aleph video-editing model
Runway has released Aleph, an in-context video model that edits existing footage through tasks such as object removal, new camera angles and relighting, and it is available now to all paid users.
Runway has introduced Runway Aleph, which the company calls a "state-of-the-art in-context video model" for multi-task visual generation.
Aleph works on an input video and performs a range of edits on it. Runway lists these:
- adding, removing and transforming objects
- generating "any angle" of a scene
- changing style and lighting
The company says the model can also do "many other tasks." Runway calls Aleph "a novel step forward for multi-task video generation and manipulation" and says it sets "a new frontier" for the field.
Aleph is available now to all paid Runway users. The research page includes a "How to Use Runway Aleph" video and a BibTeX citation that credits the work to Runway, dated 2025.
The announcement gives no benchmarks, evaluation results, architecture details, training data information or comparisons with other video models. So nothing in the release backs up the state-of-the-art claim. Runway also did not give pricing beyond saying the model is limited to paid plans, and it did not mention limits on clip length or resolution, or API access.
For builders, the main point is that Runway frames Aleph as one model that handles many editing tasks on existing footage. That differs from tools that only generate video from text or images. How well it performs will have to be judged from hands-on use until Runway publishes technical details.