A model that transfers facial expression and movement from an ordinary video recording onto another character. Record yourself on a phone, attach a frame of the hero, and the hero does the acting.

Video model

Runway AI: online video generation, an alternative to Sora.

Q: Which Runway models are available in GPTunneL?

The current list is always visible in the «Every Runway model» block above — it is built straight from the platform catalogue.

Q: Which language should I write the description in?

Whichever you like — the model understands dozens of languages, and translating the scene into English changes nothing.

Q: How is Runway different from Kling?

Kling is stronger at believable motion physics and makes sound together with the picture. Runway is closer to an editing kit: besides generation there are edits to finished video by description (Aleph) and performance transfer onto a character (Act-Two). Both families are available in GPTunneL.

Q: Can I change a video that already exists?

Yes, that is what Aleph is for: upload the clip and ask in words to swap the background, remove an object or change the time of day — nothing has to be rebuilt from scratch.

Q: Who owns the generated clips?

The rights to the result stay with you. Individual model licences may add restrictions — their terms are published by the developer.

Runway models in GPTunneL: video from text and from a frame, consistent characters, edits to a finished clip by description. Pay per second.

Try Runway Available models

What the Runway family is good at

Traits Runway keeps from one generation to the next — you can count on them whichever model you pick.

Video from text and from a frame

The model works both ways: it builds a clip from a description or brings a ready picture to life — a photograph, a concept, a still from a shoot.

The character stays the same across shots

This is what puts the family into production work: the same hero, prop and location are recognisable across different clips, so the shots cut together into one scene.

Edits to finished video by description

The Aleph model changes an already shot or generated clip on a text request — swap the background, remove an object, change the time of day.

An actor's performance on your character

Act-Two transfers facial expression and movement from an ordinary recording of an actor onto a drawn or generated character — all you have to film is yourself.

Which model to pick

Runway is four tools — for different stages of work on a clip.

	Runway 4Fast drafts	Runway 4.5The default pick	Aleph 2.0Edit what exists	Act-TwoAnimate a character
Best for	Trying ideas, draft clips	Finished scenes with characters and props	Edits to already shot or generated video	An actor's expression and movement on your character
What you feed in	A description or a frame	A description or a frame	A finished clip and a text request	A recording of an actor and a frame of the character
Strong point	Speed	Motion physics and consistent characters	Targeted changes without a reshoot	Lifelike expression
Price per second	Low	Medium	Highest	Lowest

Runway is billed per second of the finished clip. Exact per-model prices are on the Pricing

Every Runway model

Release dates follow Runway's official announcements.

May 2026

Runway Aleph 2.0

The second generation of the editing model: it handles clips up to 30 seconds at 1080p, spreads a single edit across every cut at once, and lets you set the look you want with a reference frame instead of words.

December 2025

Runway 4.5

The current generation for generation: physics got noticeably more honest — collisions, the weight of objects and human motion look filmed rather than drawn. The model holds a busy multi-object scene together and follows the prompt more closely.

July 2025

Runway Aleph

A model for editing finished video. The clip is not rebuilt from scratch — you simply ask in words to swap the background, remove an object or change the time of day.

June 2025

Runway Act-Two

Performance transfer. Record yourself on a phone and the expression and movement end up on a drawn or generated character.

March 2025

Runway 4

The generation that put the family into production work: characters, props and locations stay recognisable from shot to shot, so separate clips cut together into one scene.

Full release history

June 2024

Runway 3

A clear step up in realistic motion and in how closely the model follows the scene description. It stayed in the service as the fast, cheaper option.

June 2023

Runway Gen-2

The first widely available model that made video straight from text. Mass video generation with neural networks starts here.

February 2023

Runway Gen-1

An early model from the company — it did not draw from scratch but restyled footage you had already shot.

How billing works

There is no subscription: you top up one balance and spend it on any model on the platform.

Pay per second

You are charged for the length of the finished clip and nothing else. Not using it costs nothing: no limits, no monthly fee.

One balance for every model

Runway, ChatGPT, Claude, image and music generation — all out of the same wallet. No separate subscription per service.

Top up the way you prefer

International cards, Apple Pay, Google Pay or crypto. The minimum top-up is $5.

See pricing

How to get started with Runway

One account for every model on the platform.

Emailyou@example.comPassword••••••••

An astronaut walks across a red desert, dust rising from their boots, mountains on the horizon

5 seconds

Runway 4.5

The model switches right inside the input — you can change it between clips.

The finished video arrives at working resolution.

Try Runway in GPTunneL

Signing up takes a minute, and $5 on the balance is enough to see whether the model fits your task.

Try Runway

Frequently asked questions

/01Do I need a VPN to use Runway through GPTunneL?

No. Requests go through GPTunneL's infrastructure, so the models open like any ordinary website — no VPN and no proxy.

/02Which Runway models are available in GPTunneL?

/03Which language should I write the description in?

/04How is Runway different from Kling?

/05Can I change a video that already exists?

/06What is Act-Two?

/07Who owns the generated clips?

Runway is the American company that started mass video generation with neural networks: its Gen-2 was the first, back in 2023, to let people make clips straight from a text description. Today Runway is not one model but a small kit of tools around video.

Generation proper makes a clip from a description or brings a ready frame to life. From the fourth generation onwards the family keeps characters and objects recognisable from shot to shot — which is exactly why it goes into production work, where separate clips have to cut together into one scene. Two models stand apart: Aleph edits already shot video on a text request, and Act-Two transfers facial expression and movement from a recording of an actor onto a drawn or generated character. The model understands scene descriptions in dozens of languages, so there is nothing to translate.

Runway models are available in GPTunneL — worldwide, without a VPN, with one balance and billing per finished generation.