Hotel Lobby AI App: Making the Video From Your Phone
Two photos in, one studio performance out — and nothing to install. Updated October 1, 2026.
App or browser?
This runs in the browser, so there is no app-store listing and no version to update. That suits this format particularly well: you are uploading two photos, and most people already have them on the phone in their hand. Open the page, pick the two images from your camera roll, and go.
The same page works on a desktop browser. Files transfer faster on desktop, so if your photos are large, the upload is the only part that noticeably changes.
What to prepare before you start
- Two photos, one per subject — a selfie and a photo of a friend works
- A supported file — JPG, PNG or WebP, under 10 MB each
- Permission — both photos must be yours to use, and human subjects must be consenting adults
- Your recovery code, once your account exists. A used code is replaced with a new one, and if you lose both your session and your code, email recovery is not available
How it runs
- Upload one photo for each performer.
- Choose duration — 5 or 10 seconds — and resolution, 768P or 2K.
- Add optional direction for staging and movement, then check the credit charge shown on screen.
- Submit. Tasks keep running on the server even if you close the tab, so you do not need to keep the phone awake.
The current generator defaults to MiniMax H3 reference-to-video. Results and processing times vary, and a specific likeness or performance is not guaranteed.
What it costs
| Output | Credits |
|---|---|
| 5 seconds, 768P | 50 |
| 10 seconds, 768P | 100 |
| 5 seconds, 2K | 85 |
| 10 seconds, 2K | 170 |
Planned one-time packs are Starter — 300 credits for US$9 — and Creator — 1,000 credits for US$28. Public paid checkout is not yet open, and the final pack quantities and prices are disclosed on the pricing page. There is no subscription and no automatic renewal. Rejected prompts are not charged, and a confirmed provider failure restores the generation credits.
What it does not do
It does not replace a face in footage you already have, and it does not verify identity, age or consent. Reference photos are screened with OpenAI image moderation, and a blocked or unavailable check prevents generation — but confirming that the person in a photo agreed to appear remains your responsibility. What counts as unacceptable is listed in the Acceptable Use Policy.
Frequently asked questions
Is there an Android or iOS app?
No. The generator runs in the browser, so there is nothing to install and no app-store version to update.
Does it work on mobile data?
Yes, but uploading two photos takes longer on a weak connection. Use wifi if the files are large.
Do I need two photos?
Yes, one photo per performer. Each image should show a single clearly visible subject.
Do I lose the video if I close the tab?
No. Generation continues on the server. The result is reached through your account, but permanent availability is not guaranteed, so download anything you want to keep.
Is it free?
There is no free generation allowance at the moment. A run costs 50 to 170 credits depending on duration and resolution.
Can I post the clip on social media?
Yes. Choose the vertical 9:16 format for short-form feeds. You are responsible for having permission to use both photos.