# Live, steerable video models

These models keep generating while you steer them: a new prompt, a different source video or a changed scene takes effect without starting a new clip.

They are not navigable worlds. You direct what happens on screen rather than walking through a place.

- [EdenLive](/world-models/edenlive): Source video (game footage or a live camera), a scene prompt and a reference image. Open weights; Custom terms; Real-time, per authors.
- [Helios](/world-models/helios): Text prompts that can change while the video streams. Open weights; Permissive license; Real-time, per authors.
- [Krea Realtime 14B](/world-models/krea-realtime-14b): Text prompts and live video input. Open weights; Permissive license; Real-time, per authors.
- [LongLive 2.0](/world-models/longlive-2): Text prompts switched during generation, across multiple shots. Open weights; Custom terms; Real-time, per authors.
- [PixVerse R1](/world-models/pixverse-r1): Live prompts and chat in persistent, shared worlds. No public weights; Service terms; Real-time, per authors.
- [Visko Orbis](/world-models/visko-orbis): Text prompts and an optional start image, steered between generated chunks. No public weights; Service terms; Real-time, per authors.
- [Vivix W1](/world-models/vivix-w1): Text, voice and motion. No public weights; Service terms; Real-time, per authors.

Updated 2026-11-03. Full directory: /world-models