runway capabilities

Act-Two is Runway's AI-powered full-body motion-capture model that transfers a performer's movements, facial expressions, and hand gestures from a driving video to any animated character. No specialized studio equipment is required — a standard webcam recording suffices as the driving input.

Runway exposes a REST API and official SDKs (Node.js, Python) so developers can integrate AI video and image generation into their own applications. The API covers text-to-video, image-to-video, video-to-video, image generation, audio generation, upscaling, and real-time avatar sessions.

Runway's image generation capability (originally called Frames, now Gen-4 Image) produces high-fidelity still images from text prompts, optionally guided by one or more reference images for consistent character, scene, or style. An 'Aesthetic Range' slider (0–5) controls how far outputs deviate from the exact prompt.

Runway's Lip Sync feature (also surfaced as the 'Add Dialogue' app) animates a static portrait or video face to synchronize mouth and jaw movements with a supplied audio track or typed text, generating a short talking-head video.

Motion Brush lets users paint specific regions of an input image and define directional movement vectors for each region, controlling how different parts of the scene animate in the generated video. Advanced Camera Controls (Director Mode) let users specify camera motion type (zoom, pan, tilt, truck) and intensity for the duration of the clip.

Runway uses a credit-based billing system across five tiers: Free (125 one-time credits), Standard ($12/mo, 625 credits), Pro ($28/mo, 2,250 credits), Unlimited ($76/mo, 2,250 credits + unlimited relaxed-rate generations), and Enterprise (custom pricing). All paid plans grant commercial use rights and remove watermarks; users retain ownership of generated outputs per Runway's Terms of Use.

Runway Gen-4 and Gen-4.5 generate short cinematic video clips from a text prompt alone, or from a text prompt combined with a reference image as the first frame. Gen-4.5 (released December 2025) is the current flagship model, offering improved motion realism, character consistency, and native audio generation.

Runway's video-to-video transformation lets users take an existing video and restyle its visual aesthetic, lighting, color, or character appearance using a text prompt and/or a reference style image. The Gen-4 Aleph model (upgraded to Aleph 2.0 in May 2026) powers frame-level edits that propagate consistently across the full video clip.

Runway Gen-4 introduced 'world consistency' — the ability to keep a specific character, object, or environment coherent across multiple independently generated video clips by supplying a reference image. This enables directors to regenerate the same character from different angles and in different scenes without frame-by-frame manual continuity work.