Coverage described Solaris as treating software like video, and noted the 250-participant user study results and remaining reliability challenges
Runway Launches Solaris, a Frame-by-Frame UI Model Without Code
The approach bypasses the design-to-code translation step by rendering interface pixels directly, which Runway claims allows interfaces to be generated on demand rather than prebuilt. The system's known limitations, internal-only benchmarks, and Runway's own characterization of it as an early research system mean its practical applicability to production software remains unconfirmed.
The full picture
Runway's Solaris generates interactive application interfaces frame by frame in real time, predicting what the next screen state should look like after each click or drag rather than writing code. The system is built on Runway's Gen-4.5 video model and GWM-1 autoregressive world model, splitting reasoning from rendering. Runway targets 720p output and sub-500-millisecond per-frame latency, though these figures are not independently verified. Solaris is not publicly available; Runway is accepting early access requests and has released no weights, code, API details, or pricing.
How it developed
Runway introduced Solaris as its first Interface World Model, with early access available by request
Sources
2 more sources
Want this in your inbox?
I send a short email each morning with the stories that moved. If you would rather just read here, that works too.
Subscribe free