AI video generation/4K output
Make a short video. From text, an image, or a clip.
Start from a prompt, a still image, or a clip. Pick a model, set duration and aspect ratio, then generate in one workspace. Free credits to start.
- 22
- video models
- 4K
- on 7 of 22
- 60s
- longest clip
- 7/22
- ship with audio
[01] Ways in
Six doors, one workspace.
Every workflow below has its own page, its own model defaults, and its own set of controls. Start wherever your material already is.
Two-photo duo clip
Hotel Lobby AI Video Generator
- Start with
- Two separate portraits
- Create
- A 10- or 15-second duo clip
Image to video
Turn a photo into a short video.
- Start with
- A supported reference image
- Create
- A short motion-guided video
Text to video
Turn a prompt into a short video.
- Start with
- A written scene and motion brief
- Create
- A short generated video
Source-video transformation
Turn a clip into a new version.
- Start with
- A supported source video
- Direct with
- A written transformation direction
Product-focused video
Turn a product photo into a listing video.
- Begin with
- A product role and shot priority
- Optionally add
- An approved product reference image
Vertical 9:16 short video
Make a 9:16 video for the phone screen.
- Frame
- Native 9:16 vertical composition
- Plan for
- Reels, Shorts, and story placements
[02] The pipeline
How to make a Hotel Lobby AI video
Start with the two photos. Length, resolution, and frame have useful defaults and can be changed before you generate.
Add two authorized photos
Use one clear person or pet per image and decide who should perform on the left and right.
Choose length, resolution, and frame
Pick 10 or 15 seconds, a supported resolution, and 9:16 or 16:9. The credit quote updates before you start.
Generate and review
Play the finished clip, check both identities, motion, and sound, then download the MP4.
01 / HOTEL LOBBY
TWO PHOTOS. ONE DUO CLIP.
Upload two portraits and drop your duo into the orange booth. The scene, camera, and performance are already set.
Open the workflow →02 / IMAGE TO VIDEO
ONE STILL. THIRTY SECONDS.
Drop in a frame you already own and let the model find the movement that was implied in it.
Open the workflow →Source-video transformation
Keep the motion. Change the look.
Choose the source-video workflow when an uploaded clip provides the movement and timing you want to begin from. Add the supported video, describe the visual transformation, then inspect the generated version as a new result before reuse.
A movement-led starting point
Use an approved source clip when its movement or timing gives the next version a useful foundation.
A specific visual transformation
Explain what should change in the new version, such as scene treatment, material feeling, or overall visual direction.
A source-aware review step
Compare the result with the source clip before relying on continuity, product details, or exact visible information.
[03] The board
Twenty-two models, fully spec’d.
Showing 22 of 22 video models
[04] Controls
What every dial does.
Resolution
480p → 4K
Set per model. Only 7 of the 22 reach 4K.
Duration
1s → 60s
Some models expose fixed lengths, others a full range.
Aspect ratio
9 max
Chosen at generation time, not cropped afterwards.
Audio
7/22
Generated on the same track as the picture.
Prompt length
30k
Long enough for a full shot description, not a slogan.
Start/end frames
5/22
Pin where the shot begins and where it lands.
[R] Framing
Framed for where it lands.
Pick the aspect ratio at generation time instead of cropping a 16:9 master afterwards and losing the composition. Support varies by model — the count under each frame is the current one.
9:16
Shorts, Reels, TikTok
18/22 models
1:1
Feed and marketplace
14/22 models
16:9
YouTube and web
18/22 models
21:9
Cinematic and hero
6/22 models
22
video models
60s
longest single clip
4K
top resolution
7
ship with audio
5
take start/end frames
[05] Made here
Twelve, playing at once.
[W] Why here
Twenty-two subscriptions, or one.
01
One account, every model
Seedance, Kling, Veo, Wan, Grok, MiniMax and the rest sit behind the same prompt box and the same credit balance. No twenty subscriptions, no twenty dashboards.
02
Limits published up front
Every model page lists its resolutions, durations, framing options and whether audio comes with it — before you spend a credit finding out.
03
The whole take is kept
Prompt, reference, model and settings stay attached to every result, so a direction that worked can be returned to instead of reconstructed.
[L] Before you publish
Before you generate a Hotel Lobby video
01
Upload people who agreed
Use original photos you have permission to publish. Do not upload children or identities you do not control.
02
Audio follows the prepared performance
Sound is generated with the template. This workflow does not promise word-level lip-sync or the original track.
03
Review the final clip
Faces, hands, clothing, positions, and timing can drift during generation.
[C] Catalog
Read before you generate.
[$] Credits
One balance, every model.
Every new account starts with 50 free credits. Video models cost more than image models — the current price is shown before you generate.
Free
Get started
$0
No payment method required
50 welcome credits
- 50 welcome credits at sign-up
- Start with available models
- No subscription required
Starter
For occasional projects
$9.90/ month
Billed monthly. Cancel before the next renewal.
1,000 credits every month
Regular rate: approx. $0.99 per 100 credits
- Use credits for image and video generation
- Choose any available model and output settings
- Commercial use
- Unused subscription credits roll over and remain available.
Pro
For frequent work
$19.90/ month
Billed monthly. Cancel before the next renewal.
2,000 credits every month
Regular rate: approx. $0.99 per 100 credits
FREE GIFT
+200 credits · one-time on your first subscription
First month total: 2,200 credits
- Use credits for image and video generation
- Choose any available model and output settings
- Commercial use
- Unused subscription credits roll over and remain available.
Ultimate
For continuous production
$49.90/ month
Billed monthly. Cancel before the next renewal.
5,000 credits every month
Regular rate: approx. $1 per 100 credits
FREE GIFT
+500 credits · one-time on your first subscription
First month total: 5,500 credits
- Use credits for image and video generation
- Choose any available model and output settings
- Commercial use
- Unused subscription credits roll over and remain available.
Need extra credits?
Add a one-time credit pack whenever you need more capacity. These credits can be used alongside a subscription and never expire.
Credits roll into one balance and work across every image and video model.
[F] Frequently asked
Asked before you ask.
Every answer here is the same one on the workflow page it came from, so nothing on this page contradicts the documentation.
It is a two-photo video generator for the orange booth and shared performance setup. Upload two permitted photos and the tool creates a short AI duo clip with those two performer roles.
Sign in, upload one photo for each person, choose resolution, duration, and frame, then generate. The finished MP4 appears on the page and in My videos.
You can start from an image when you choose a video model that accepts an image reference. The supported formats, limits, and output controls are shown by the active model.
Focus on the change over time: action, camera movement, timing, environment, and visual feeling. The image already provides the starting appearance, so the prompt should explain how it should develop.
A text-to-video generator creates a short video from a written scene direction. In ImagineClip, the prompt and selected video model determine which inputs and output controls are available.
Start with the subject, action, setting, and camera direction. Add timing, light, atmosphere, or style only after the core motion is clear.
A video-to-video workflow starts from a supported source clip and uses a written direction to create a new generated version. The result should be reviewed as a new video, not as an exact edit of the source.
Yes. ImagineClip includes a source-video workflow for compatible video models. Upload requirements and available output controls are shown in the active workspace.
It is a product-focused way to plan a generated short video. In ImagineClip, the result is shaped by the prompt, selected video model, and any compatible reference material you add.
Yes, when the selected video model supports image references. Use an approved source image and review the model-specific upload limits in the workspace.
It creates video in a native vertical frame for phone-first placements. ImagineClip provides compatible 9:16 model controls while final captions and publishing details remain separate.
Kling 3.0, Seedance 2.5, Wan 2.7, and several other current models expose 9:16. Duration, sound, reference inputs, and quality still vary by model.