Happy Horse 1.0 is now on ModelsLab

Try Now
Skip to main content
Available now on ModelsLab · Video Generation

Runway Aleph — Video-to-Video Editing via APIEdit any video with a text prompt. Swap subjects, restyle scenes, change camera angles.

Try Runway AlephAPI Documentation

Gen-4 Aleph — RunwayML's Video Editing Head

Video-to-Video

Instruction-Based Video Editing

Describe the edit in plain English — swap a subject, restyle a scene, change camera angle — Aleph applies it while preserving the source video's motion and timing.

Gen-4 Architecture

RunwayML's Newest Editing Head

Built on Runway's Gen-4 foundation and purpose-trained for editing workflows, Aleph handles subject swaps, object insertion, style transfer, and camera reframing in a single prompt.

Temporal Coherence

Consistent Across Frames

Maintains identity, lighting, and spatial structure across every frame — no flicker, no morphing, no manual keyframing.

One API Key

No RunwayML Subscription

Call Aleph through the same ModelsLab REST API you already use for Kling, Seedance, Wan, and other video models. One key, one bill.

Async Friendly

Webhook Callbacks

Pass a webhook URL — ModelsLab POSTs the edited MP4 to your endpoint when processing completes. Essential for batch jobs and production pipelines.

SDKs Included

Python & JavaScript

Official SDKs wrap every endpoint including Aleph. Or call the REST API directly — we publish an OpenAPI spec for auto-generated clients.

Examples

See what Runway Aleph can edit

Copy any prompt below and try it yourself in the playground.

Ad Creative Variation

“change the product in frame from the original bottle to a red can of sparkling water, keep the same motion, lighting, and background”

Style Transfer

“apply a cinematic film noir look — high contrast black and white, soft vignette, dramatic shadows”

Scene Replacement

“replace the background with a neon-lit Tokyo street at night, preserve the subject and their motion exactly”

Camera Reframe

“reframe this clip as a low-angle tracking shot instead of the current eye-level composition”

For Developers

A few lines of code.
Upload a clip. Describe the edit. Receive the edited MP4.

ModelsLab handles the infrastructure: fast inference, auto-scaling, and a developer-friendly API. No GPU management needed.

  • Serverless: scales to zero, scales to millions
  • Pay per clip, no minimums
  • Python and JavaScript SDKs, plus REST API
API Documentation 
import requests


response = requests.post(
    "https://modelslab.com/api/v7/video-fusion/video-to-video",
    json={
  "key": "YOUR_API_KEY",
  "prompt": "change the car colour to cherry red, keep everything else identical",
  "model_id": "gen4_aleph",
  "init_video": "https://example.com/source.mp4",
  "num_frames": 120,
  "output_type": "mp4"
}
)
print(response.json())

FAQ

Common questions about Runway Aleph — Video-to-Video Editing via API

Read the docs

Runway Aleph is RunwayML's Gen-4 video-to-video editing model. It takes an existing video clip plus a natural-language instruction and returns the same clip with the edit applied — subject swaps, style transfer, camera changes, object replacements — while preserving the source video's motion and timing.

Gen-3 is primarily text-to-video (generate a clip from a prompt). Aleph is video-to-video (edit an existing clip from a prompt). Aleph is purpose-built for editing tasks where preserving the original motion matters — ads, film pre-viz, product demos — whereas Gen-3 is for generation from scratch.

No. ModelsLab routes Runway Aleph through a single REST API — you only need a ModelsLab API key. No separate RunwayML account, no separate billing, no quota management.

Subject and object swaps, style transfer (e.g. "make it anime", "cinematic film noir"), scene changes, camera angle shifts, lighting adjustments, and object additions. Anything describable in a short natural-language instruction that preserves the source video's structural layout.

Per API call, Aleph handles 5-10 seconds of video. For longer edits, break the source into segments, edit each one, and stitch them together. Most production pipelines chunk at scene boundaries for consistency.

Pay-per-generation pricing with transparent per-clip rates published on the pricing page. No monthly commitments, no credit systems. Every ModelsLab account gets free credits to test Aleph before picking a plan.

Use model_id 'gen4_aleph' at the /api/v7/video-fusion/video-to-video endpoint. Full request/response schema, sample payloads, and SDK snippets live at docs.modelslab.com.

Yes. Pass a webhook URL in your request body and ModelsLab POSTs the completed edited video URL to your endpoint when processing finishes. Essential for long-running edits and batch workloads.

Yes. Most production pipelines combine models: generate a base clip with Gen-3, Kling, or Wan 2.7, then edit it with Aleph. One API key, one billing account, one SDK — route between providers based on use case.

Yes. The ModelsLab Python SDK wraps every video endpoint including Aleph. You can also call the REST endpoint directly with the `requests` library — we publish an OpenAPI spec so any autogenerated client works in any language.

Ready to create?

Start generating with Runway Aleph — Video-to-Video Editing via API on ModelsLab.

Try Runway AlephAPI Documentation