Introduction

Runway Aleph introduces powerful generative capabilities that enable the editing and manipulation of existing footage. Learn more here: Introducing Runway Aleph

This core functionality introduces a wide variety of new applications and workflows with generative technology in storytelling, such as:

• Showing new angles of existing video scenes
• Editing specific subjects, objects, or environments within footage
• Adding VFX to footage
• Changing the lighting source of existing footage

These are just a few examples, as we are discovering new applications each day. See our Aleph Prompting Guide documentation for a list of examples. We will add more use cases as discoveries are made.

Article Highlights

• We recommend using Chat Mode as your creative partner when creating with Aleph
• Aleph supports a maximum duration of 5 seconds for a single generation
• Aleph automatically crops videos that don't scale to supported resolutions
• Prompt by including an action verb and description of the change

Spec Information

Step 1 — Accessing Aleph

Aleph is available in both Chat Mode and Tool Mode within a generative session. 

We recommend Chat Mode for Aleph because it acts as a creative partner. This collaborative approach makes it easier to explore new ideas and refine your vision through conversation.

Tool Mode can be helpful in cases where you'd prefer a more manual approach, such as granular control over exact prompts when iterating.

To access either mode, begin by starting a new session or opening an existing session.  

• Chat Mode — Click the Chat Mode button and upload a video to the prompt.
• Tool Mode — Ensure the Gen-4 model is selected, then upload a video to the prompt canvas. 

Before uploading, keep in mind that Aleph defaults to using the first 5 seconds of your video and automatically crops videos that don't match supported resolutions. For longer videos, you can select which portion to use after uploading with the trimming interface.

Step 2 — Drafting the prompt 

Tip: For more prompt and output examples, please see Aleph Prompting Guide

There are no strict rules for prompting with Aleph, but simplicity is encouraged. Effective prompts include two key elements: 

• Use of an action verb that describes the scope of changes
• Description of the target transformation

Examples of action verbs include add, remove, change, replace, re-light, and re-style. 

The examples below demonstrate different possibilities with this core approach:

Prompt	Input	Output
Add ice that starts spreading over the hand
Change the camera to a wide angle that shows the entire person
Change the season in the original video to winter. Snow and ice on the road
Add fireworks to this scene

Using input images

You can accompany your input video with an image to influence color, styles, lighting, and much more.

When using an image, specify how your image and video inputs should interact. In example, you might try prompts such as:

• Change the color of the house in the video to the color from the image
• Re-light the video using the lighting from the image
• Restyle the video using the 3D animation style from the image

Prompt	Input	Output
Change the color of the house in the video to the color from the image

Please note that subjects and objects from images may not be consistent at this time. 

For elements that must remain consistent, we recommend ensuring that they are included in your input video. You can generate consistent subjects and objects in References and then animate the image with Gen-4.

Step 3 — Generating and Iterating

After drafting your prompt, send the message to Chat Mode or click Generate in Tool Mode to start the generation.

Once the generation is complete, you can ask Chat Mode to make further changes to the prompt to iterate or upscale the result to 4K.