The Hotel Lobby AI prompt, already built in.

A familiar studio look, from two photos

The Hotel Lobby AI template is built around a seamless orange performance space and a single hanging microphone shared by two subjects. Our server applies the preset whenever you create a video; there is no prompt field to fill in.

Bring the two reference subjects into an orange studio. Place one on the left and one on the right, with a suspended microphone between them. Use lively duet movement: alternating rap lines, expressive hand gestures, a playful turn and shared reactions, with both performers in view.

This is a readable summary of the preset’s direction. Your images provide the cast, while our fixed example video guides the scene and performance. The playable example on our homepage was newly generated from a reference performance and two performer images. Your results will vary.

Each photo has its own role

Photo 1 is the appearance reference for the left performer; Photo 2 is the reference for the right performer. The prompt asks the model to keep their recognizable features and visual styles separate. It also asks for consistent outfits and no additional performers.

This same setup works as a starting point for two friends, a person and a pet, two pets or character art. You do not need to select a subject category. Results depend on how clearly the model can interpret each reference.

Expressive gestures, shared focus

Your two photos and our fixed example video guide the performers, moves and scene in a new performance. The prompt calls for alternating lines, bouncing shoulders, pointing, a turn by the right performer and a brief shared finger-to-lips gesture. Medium-wide and closer two-shots keep both performers visible, with consistent lighting, backdrop and microphone. Short outputs focus on the opening exchange.

It also discourages merged features, sudden transformations, captions and logos. These are directions to a generative model, not guarantees that every frame will follow them perfectly.

What the template does—and what to check

The studio creates a new video in 9:16 portrait (default), 16:9 landscape or 1:1 square, with your choice of 5, 10 or 15 seconds and 768P or 2K quality. The reference video guides the choreography and shot changes; it does not guarantee an exact match, a particular music track or synchronized lip movement. Watch the completed result to check appearance, motion and any audio before sharing.

Searching for a Migos AI video generator often leads to this orange-studio duo format. Try Hotel Lobby is an independent tool, using a reference-guided scene. It is not affiliated with the performers or original recording.

Your photos are the most useful control

Choose a clearer portrait to give the model more facial detail. Leave space around the subject to improve the framing reference. Use Swap photos to reverse the left and right roles. There is no need to find a “magic” prompt before getting started.

For file requirements and a complete walkthrough, read how to make a Hotel Lobby AI video. All generations use one-time video credits.