Runway launches Media Router to pick AI models for images, video, and audio

In short: Runway has launched a new tool that automatically chooses which AI model to use when a company wants to generate an image, video, or audio clip.

What happened

Runway announced Runway Media Router, a new feature available through Runway Dev, its developer platform launched earlier this month. Runway Dev lets companies connect to Runway’s tools through an API, which is a way for software to talk to other software (like plugging one app into another).

The Media Router is designed to pick the best AI model for a specific request. A developer can tell it what matters most, like higher quality, faster results, or lower cost. It can then route the request to Runway’s own models or to third-party models for images, video, and audio.

Runway said model routers are common for text chatbots, but this one is built for “generative media,” meaning AI that creates pictures, videos, or sound. Runway also said the number of media-making models has grown quickly, which makes it harder for teams to keep testing each new release.

Runway’s product leaders said the router uses the company’s internal experience judging outputs, such as how video models handle motion or how voice models handle lip syncing. They also said developers may set preferences, for example avoiding certain model providers based on business or policy concerns.

Why it matters

For people using products from companies like Adobe or Expedia, this could mean more AI features show up inside the apps they already use, without needing to visit Runway’s site. It is like having a switchboard operator (a helper who connects your call to the right person) that chooses the best tool behind the scenes based on price, speed, or results.

Source: TechCrunch AI