Real-time video
Runway Characters
Turn one image into a live conversational video character.

Real-time video
Turn one image into a live conversational video character.
Product insights
Runway Characters streams expressive conversational video at 24 FPS, with 37 ms effective model time per frame and about 1.75 seconds of server-side response time.
Traditional video generation often makes you wait for a finished clip. Real-time video products bring feedback and control back into the generation loop.
For creative technologists, developers, and product teams building live visual experiences.
Hosted access is available through the official website. Plans, credits, regions, and API terms vary by product.
Runway Characters creates real-time conversational video agents from a single reference image. Characters listen, respond, gesture, and stream video at 24 FPS during a live session.
Start with one reference image in any supported visual style.
Assign a voice, knowledge source, and optional tools to the character.
Talk to the character directly or embed it through the SDK, widget, or LiveKit integration.
Runway reports 24 FPS streaming, 37 ms effective model time per frame, and about 1.75 seconds from the end of user speech to the start of the character’s response.
Official website snapshot
runway.com
Runway Characters streams expressive conversational video at 24 FPS, with 37 ms effective model time per frame and about 1.75 seconds of server-side response time.