---
title: "Runway's Aleph 2.0 Now Lets Developers Swap Any Video Detail Without Reshooting"
subtitle: "Runway's Aleph 2.0 is now on the API: edit up to 30 seconds of 1080p multi-shot footage with a single prompt, changing only what you ask for"
url: "https://alphasignal.ai/news/runway-s-aleph-2-0-now-lets-developers-swap-any-video-detail-without-reshooting"
author: "AlphaSignal Newsroom"
author_url: "https://alphasignal.ai/editorial-team"
company: "Runway"
company_url: "https://alphasignal.ai/company/runwayml"
published: "2026-06-02T19:41:28.000Z"
updated: "2026-06-16T20:29:00.401Z"
type: "News"
topics:
  - "Video"
  - "Api"
subtopics:
  - "Video Editing"
upvotes: 247
read_time: "3 min"
source_url: "https://dev.runwayml.com/"
comments: 0
---

# Runway's Aleph 2.0 Now Lets Developers Swap Any Video Detail Without Reshooting

> Runway's Aleph 2.0 is now on the API: edit up to 30 seconds of 1080p multi-shot footage with a single prompt, changing only what you ask for

Most AI video tools start from nothing. You write a prompt, the model invents footage, and you iterate until something usable comes out. [Runway's Aleph 2.0](https://runwayml.com/news/introducing-aleph-2-and-edit-studio) flips that entirely. It takes footage you already have, applies one targeted change, and hands back the same clip with everything else untouched. Now that model is available through the Runway API, meaning you can wire it directly into your own products and pipelines.

## The problem it actually solves

Commercial production starts from footage that's already been shot and color-graded, that now needs a single specific change. The seasonal variant of one campaign film. The wardrobe swap a client signs off on three days before launch. The old answers were a roundtrip through compositing software or a reshoot. Both are slow, and both lock the rest of the clip into whatever the first edit sets in motion.

Marketers, agencies, and small businesses rarely need one impressive demo clip. They need campaign variants, product swaps, localization, aspect-ratio changes, seasonal edits, and fixes after the shoot. The value is in shortening the loop from existing asset to shippable variant.

## What Aleph 2.0 can actually do

Aleph 2.0 is Runway's upgraded flagship video editing model for transforming existing footage while keeping the rest of the clip stable. It is built for precise localized edits, single-frame visual guidance, and consistent edits across cuts and scene changes, making it useful for product swaps, packaging updates, background changes, relighting, scene cleanup, and other post-production workflows that need the output to stay close to the source video.

There are four core capabilities in this release:

- **Localized edits:**
  Describe what you want to change , a product color, a hairstyle, a piece of clothing , and Aleph 2.0 changes only that. The background, lighting and other details stay just as they were.
- **Image-guided edits (frame anchors):**
  Edit one frame and Aleph 2.0 modifies the rest of your video to match , preserving everything you didn't ask to change.
  This is the key unlock for brand-precise work: pull a still, edit it in Photoshop or any image tool, then send it back as a reference. The model carries that exact look through every frame.
- **Multi-shot propagation:**
  In videos with multiple cuts or scene changes, Aleph 2.0 will apply your edit across relevant shots. You no longer need to edit your videos shot-by-shot.
- **Longer clips at higher resolution:**
  You can now edit clips up to 30 seconds long at 1080p. Long enough for ads, social posts and short-form content.

## The frame anchor is the real unlock

Text prompts work well for simple changes. They break down the moment you need something precise , a specific brand stripe, exact typography, a packshot that has to match a print campaign. That's where the frame anchor workflow comes in.

The idea: extract a still from your source clip, edit it externally in any image tool to show exactly what you want, then pass it back to the API pinned to a specific timestamp. The model treats that edited frame as ground truth and propagates the look through the rest of the footage. You can pin up to five anchor frames per clip, which lets you define how a look should evolve over time , say, a daytime scene transitioning to night across a 20-second cut.

Here's the minimal API shape for a frame-anchored edit:

```
[
  {
    "taskType": "videoInference
```

## Sources

- [X post by @runwayml](https://x.com/runwayml/status/2061895998545244342)