All news

Runway Launches Media Router to Auto-Select Video, Image, and Audio Models

Runway's new Media Router picks the best generative media model for each request based on cost, quality, or latency, spanning its own Gen-4.5 and Aleph 2.0 alongside third-party models like Seedance and GPT Image 2.


Runway launched Media Router on July 23, a routing layer through its Runway Dev platform that automatically picks which video, image, or audio generation model handles a given request instead of making developers choose one manually.

  • Developers set a config ID with cost caps (USD per second of video or per image), quality/latency preferences, and provider allow/deny lists, then call one endpoint
  • Responses include metadata showing which model was selected and why; a dry-run flag previews the choice without incurring cost
  • First-party models in the pool: Gen-4.5, Aleph 2.0, Act-Two; third-party: Seedance, GPT Image 2, ElevenLabs
  • Runway Dev customers already include Adobe, Cloudflare, ElevenLabs, Expedia, Shutterstock, and Quora
The routing really fits into that overall promise of being the easiest one-stop shop for developers. - Anthony Maggio, Runway CPO

The move pushes Runway further from being just a model maker toward an infrastructure layer for generative media, competing on selection logic rather than a single model's output. Compare image and video generations yourself in the arena.

Based on: TechCrunch · Runway
More AI news in Polish at nowosci.ai