Full Recipe for the Viral Hotel Lobby Migos Video Swap: Reference Frames + Tagged Prompts
marupelkar · reddit · 2026-10-01
After testing nearly every model, the author shares a reproducible workflow for the viral hotel lobby Migos person-swap video, open-sourced as a recipe repo: cut the source video, prepare full-body and close-up images per person, run Minimax H3 Max on Fal.ai with references, and generate 5-8 replacement frames at different timestamps with nano banana pro to guide the video. Prompting needs an LLM: clearly tag which image is who (#image1/#image2/#image3), mark close-up vs full-body, state approximate timestamps for reference frames, and don't restate motions already in the source video. Known failure modes: only one person gets swapped (re-run with output as input to reinforce the other), longer videos break edits more often, and broken audio is fixable with a simple ffmpeg command. Expect multiple generations; a paid service at vaaya.ai is also offered.
More from Multimodal
- Testing Qwen Image 2.1 and Minimax H3 on anime-to-photo conversion and film photo restoration — Ant_6431 · 2026-11-16
- Second take of a 'Cubee run' video generated with Minimax H3 — JahJedi · 2026-11-16
- Non-artist builds 14-chapter hand-painted storybook game with Fable 5.2 in 4 hours — jradoff · 2026-11-16
- Subject LoRAs on Minimax H3 Image learn poor likenesses while identical datasets excel on Qwen 2.1 — Ad_Organic · 2026-11-16
- Seedance-generated anime train scene stuns Reddit with cinematic quality — Ok-Solution-7889 · 2026-11-16
- AI-generated 'Men in Black' reimagined in 1977 goes viral — techhalla · 2026-11-16