# Reverie

- **Event:** [fal x Sequoia 72-Hour Video Hackathon](https://cerebralvalley.ai/e/fal-x-sequoia-virtual-hackathon)
- **When:** Jul 17 at 12:00 PM – Jul 19 at 8:00 PM (PDT)
- **Where:** Online
- **Team:** [dylan merigaud](https://cerebralvalley.ai/u/dylanmerigaud)
- **GitHub:** https://github.com/DylanMerigaud/reverie
- **Demo video:** https://youtu.be/lhAVPWUvQP0
- **Gallery:** https://cerebralvalley.ai/e/fal-x-sequoia-virtual-hackathon/hackathon/gallery
- **Page:** https://cerebralvalley.ai/e/fal-x-sequoia-virtual-hackathon/hackathon/gallery/16

Everything you watch has one look: the one it shipped with. Reverie lets you change it to match your taste, your mood, whatever you feel like today. Point it at any video and it repaints every frame into a new world while it plays: comic book, watercolor, anime, oil painting, film noir. Type a sentence and the whole look shifts as you watch.

Under the hood, Reverie is the real-time layer on top of the model, not the model. It runs on fal with Decart Lucy 2.5 over a WebRTC bridge, detects a video's cuts and pre-warms a pool of live sessions so the style hands off cleanly at every hard cut, and re-steers the moment you type a new prompt, all with the key server-side.

More of what fills a screen is generated now, not just recorded, and that share only grows. This was never only about movies: recolor your feed like a graphic novel, drop your game into a Ghibli afternoon, turn last night's call into a memory that looks the way it felt. Everything on your screen, finally yours.

---

Markdown version of https://cerebralvalley.ai/e/fal-x-sequoia-virtual-hackathon/hackathon/gallery/16. Site index for agents: https://cerebralvalley.ai/llms.txt · full text: https://cerebralvalley.ai/llms-full.txt