Generative AI company Runway has released Runway Media Router, claimed as the world’s first model router for generative media models.
Model Router is a new capability in Runway Dev that automatically selects the best video, image or audio model for every request. Until now, the company says, media has never had an equivalent to intelligent routing for LLMs, largely because video, image and audio models vary so much in what they can actually produce.
Runway says its Media Router solves this by matching a request to a model’s capabilities and modalities, then optimising across the qualifying models for the cost, quality and latency preferences set by the user.
Historically, every generation request meant hand-picking a model and manually updating that choice every time a new model launched, pricing shifted or users switched to a different medium. Model Router removes those steps, letting users configure what “best” means for their use case once and making it easier to evaluate and enable new models as they are released.
The Problem With Picking Media Models Manually
According to the company, Runway Dev’s core value is giving users access to the world’s best media models on one enterprise-grade platform.
“That means first-party access to Runway’s frontier models, including Gen-4.5, Aleph 2.0 and Act-Two, alongside the best third-party models, including Seedance, GPT Image 2 and ElevenLabs,” it says. “However, having every new model available on day zero has also become a source of friction. Builders tell us they do not want to become part-time model researchers, tracking which model is fastest or best suited to a given style and reconsidering that choice every time the catalogue changes. We often see teams pick a model once and stick with it long past the point at which it is the right choice.
“At enterprise scale, this gap compounds. A model that is slightly more expensive or slower may not matter for a single request, but, multiplied across thousands of generations, it becomes a real cost and latency burden that blocks scaling. Smart routing is well established for enterprise LLM usage, but we think it matters even more for generative media, where quality is a far more complex dimension than it is for language.”
How Model Router Works
Model Router replaces manual model selection with three steps:
- Set your preferences. In the Runway Dev portal, select an optimisation preference covering cost, latency and quality. Name your configuration, set a price cap, allow or deny different models or providers and optimise between your preferences. Save it once as a reusable configuration tied to a specific workflow, such as a fast in-app preview or a paid final export.
- Call one endpoint. Send your generation request to the Model Router endpoint with your configuration ID attached. There is no need to specify a model.
- Get the best model. The router filters out any model that fails your hard constraints, including capability fit, allow and deny lists and the price cap, then scores the remaining models against your cost, quality and latency preferences. It generates content with the highest-scoring model and returns the result, along with metadata identifying exactly which model was used and why.
The initial configuration dimensions available today are:
- Price cap: a hard ceiling per generation, in US dollars per second for video or per image for image models.
- Allow and deny lists: provider- and model-level rules that restrict routing to approved options.
- Set preferences: define your preferences across cost, quality and latency optimisation.
Runways says if a request cannot be satisfied by any model under user constraints, Model Router returns an explicit error naming the constraint that emptied the pool, rather than silently downgrading user settings. Before a configuration is put into production, a dry-run flag lets users validate exactly which model it would select and why, with no cost incurred and nothing generated.
Visit https://dev.runwayml.com
