Runway AI is a New York company and a creative platform for generating and editing video, images and audio with AI. Its Gen-4 model makes 5 or 10 second clips from an image plus a text prompt. The detail people miss: the starting image is required, so the prompt should focus on motion.
If you've seen an AI-made film clip or a slick product video and wondered what made it, there's a fair chance Runway was involved. It's one of the best-known names in AI video, but the name covers more than one thing, which is why searches for it get confusing.
Here's a straight explanation of what Runway is, what you can make with it, how its video tools work in practice, and what to watch for when you use it for a business. It sits alongside my broader look at generative AI video tools for business.
What Runway is and who makes it
Runway is the name of both a company and its products. According to Wikipedia, the company was founded in 2018 by Cristobal Valenzuela, Alejandro Matamala and Anastasis Germanidis, who met at NYU Tisch School of the Arts, and it's based in New York City. Runway's own homepage describes its mission as building foundational Real-World Intelligence that can understand, simulate and act in the world.
For most of us, though, Runway means its creative platform. The homepage describes Runway Creative as an all-in-one cloud platform for generating and editing video, images and audio in one workspace. That's the part this guide focuses on.
The parts of Runway people mix up
Runway's website now lists several product areas, and searchers often blur them together. Here's how they're described on Runway's own pages:
| Part of Runway | Description |
|---|---|
| Runway Creative | The workspace for making and editing video, images and audio |
| Runway Dev | An AI media platform for developers who want to build Runway's models into their own apps |
| Runway Robotics | A toolkit for robotics teams, powered by a world model |
| GWM-1 | Runway's General World Model, built to interact with the real world |
| Gen-4.5 | Runway's video model, described as having state-of-the-art motion quality and prompt adherence |
If you're a business owner or creator, Runway Creative is almost certainly what you want. The others are aimed at developers and research teams.
What you can make with Runway
The product page lists a spread of tools. A few that matter most for everyday users:
- Video generation: make clips from text and images using Runway's video models.
- Transform Video: change a scene's lighting, restyle a shot or subject, or add and remove elements using text instructions.
- Aleph 2.0 and Edit Studio: edit one frame of a video and have the change applied across the rest.
- Agent: make a video through a conversation rather than clicking through tools.
- Single-purpose apps: small tools such as Remove from Video and Upscale Video.
The page also lists image models and tools. Names on that list come and go quickly, so check the live product page before you plan around any one of them.
How Runway turns an image into a video
Runway's help centre has a detailed article on its Gen-4 video model. It's a good example of how the platform thinks, because it shows what you give it and what you get back.
- Gen-4 creates videos in 5 or 10 second durations.
- It works from an input image and a text prompt you provide.
- The input image is required on both Gen-4 models.
- The image becomes the starting point and acts as the first frame of the video.
That last point is the one people trip over. Many beginners expect to type a sentence and get a film, but Gen-4 wants a starting picture. Because the image already carries the subject, composition, colours, lighting and style, Runway's help says your text prompt should be almost entirely about the motion you want.
How to write a good Runway Gen-4 prompt
Runway publishes a prompting guide for Gen-4, and its advice is refreshingly simple. Here's the gist, as the guide states it:
Want AI doing the heavy lifting in your marketing?
I build the systems that handle the boring 80 percent, so you get your week back. Done properly, with the human kept in.
- Start with a simple prompt that captures only the most essential motion in the scene.
- Add one new element at a time, such as subject motion, camera motion, scene motion or style descriptors, so you can see what helped.
- Use positive phrasing. Describe what should happen, not what to avoid, because negative prompts aren't supported and can give unpredictable results.
- Use a high-quality input image free of visual artifacts.
- Refer to people and things in general terms, like "the subject", rather than long descriptions.
If you're new to AI video, the clips are short, so think in moments: a slow camera push on a product, steam rising from a cup, a gentle pan across a shopfront. For ideas on free ways to practise first, see my guide to free AI animation tools for beginners.
How a small business might use Runway day to day
- Turn a product photo into a short looping clip for social media.
- Restyle or relight footage you've filmed yourself, instead of reshooting it.
- Remove a distracting object from your own video.
- Mock up a storyboard or concept clip before spending time on a full shoot.
- Make a short background visual for a presentation or web page.
It works best as a way to speed up ideas, not to replace filming everything. Short clips are easy to make; a long, consistent story still needs planning, editing and your own judgement.
Runway safety, privacy and good practice
Runway's safety page says it uses AI classifiers to screen inputs and outputs for harmful content, has human reviewers for reports and appeals, and uses C2PA so AI-generated content can be traced back to its source. Its data security page says customer content is never used for testing new features or changes. It doesn't spell out model training there, so read Runway's Privacy Policy and Trust Center if that matters to you.
- Only upload images and video you own or have permission to use.
- Don't make clips of real people without their consent, and check Runway's usage policy on likenesses.
- Check Runway's terms before using output in client or advertising work.
- Tell your audience when a clip is AI-generated, especially if it could be mistaken for real footage.
If you're weighing where video fits in your marketing, that's the kind of question I help with as an AI consultant.
Frequently asked questions
What is Runway AI used for?
Runway is used to generate and edit video, images and audio. Its creative platform lists tools such as video generation, Transform Video, Edit Studio and an Agent that makes videos through conversation.
Who makes Runway?
Runway is made by Runway AI, Inc., a New York company. Wikipedia says it was founded in 2018 by Cristobal Valenzuela, Alejandro Matamala and Anastasis Germanidis.
Do I need an image to make a video in Runway?
For Gen-4, yes. Runway's help centre says the input image is required on both Gen-4 models and acts as the first frame of the video.
How long are Runway Gen-4 videos?
Runway's help centre says Gen-4 creates videos in 5 or 10 second durations.
Is Runway the same as Stable Diffusion?
No, but they're linked. Wikipedia says Runway co-released the Stable Diffusion model in 2022 with a research group at LMU Munich. Runway's own platform is a separate product.
