Made with AlreadyRan
Real results from the generator, with the prompts that made them. Click one to watch it full size.
Overview
Text to video, image to video and video with sound in one place
AlreadyRan's AI video maker brings 32 video models from different makers into one generator. Write a prompt, start from a still image, or guide the shot with reference images, video and audio — then compare the models side by side and pay for what each clip actually uses.
From text, from an image, or from references
Text to video works from the prompt alone and is the quickest way to explore an idea. Image to video starts from your picture as the first frame, so the subject, the framing and the colours are fixed and the model only has to invent the motion — the most predictable way to get the clip you imagined.
Reference to video goes further: some models take several images, video clips or audio files and use them for a character, a product, a movement or a soundtrack. Edit and extend modes work on a video you already have. The model card lists which of these each model offers.
Write the prompt like a shot list
Start with the format and the camera (documentary footage, selfie vlog, locked-off security camera), then the subject, then what happens. For longer clips, give the beats with timings — 0–3s, 3–7s — so each moment has room.
Put spoken lines in quotes and name the sounds you want. Describe a person once, precisely, and ask for them to stay identical across shots. Short, concrete words work better than long lists of adjectives.
Renders keep going when you leave
Video takes longer than an image. Generations run on our servers, so you can switch to another page, start the next take or close the tab; the finished clip waits in Recent and in your History.
Use cases
What you can make with AI video
Short films and multi-shot scenes
Write the scene as numbered shots with timings, and models that support multi-shot output cut between them in a single clip — useful for trailers, story beats and micro films.
Prompt Three-shot romantic micro short film on a rainy night in a big city. SHOT 1 (0–4s): a young man in a soaked suit runs into a bus shelter and finds a yellow umbrella hanging there with a paper tag tied to the handle. SHOT 2 (4–8s): close-up of the tag in his hand…
Characters who talk to camera
Put the lines in quotes and say how they are delivered. Models with sound generate the voice with the picture, which suits vlogs, sketches and social clips that need a punchline.
Prompt Selfie-style vlog filmed at arm's length: a red-bearded Viking on a longship in the stormy North Sea, rain dripping off his helmet, crew rowing behind him. He talks to the camera like a tired travel influencer: “Day thirty-four. Still no land. Bjorn ate the last fish. Again.”…
Action and sports footage
Describe the camera as well as the subject — tow-in, drop, barrel, exit — so the model knows where the shot goes. Fast motion is where models differ most, so compare a few on a short take first.
Prompt Big-wave documentary footage: a surfer in a bright yellow wetsuit drops into a monstrous 20-meter wave at a remote reef. BEATS: 0–3s jet-ski tow-in, she lets go of the rope and rises to her feet; 3–7s she plunges down the face as the lip throws overhead…
Animation in any style
Name the style first — 2D anime, claymation, stop motion — then the character and the gag. A short, specific character description keeps the look steady from the first frame to the last.
Prompt 2D anime comedy short starring an original character: Pip, a clumsy apprentice witch with a too-big pointy hat and freckles, trying to flip pancakes with magic in a cozy cottage kitchen…
Formats people recognise
Security-camera footage, news interviews, home video, heist-movie HUDs: naming a format sets the framing, grain and on-screen text in one phrase, and gives a clip its joke or its realism.
Prompt Backyard security-camera footage at night: fixed high angle, grainy infrared black-and-white, timestamp “CAM 2 · 03:12:44”. A family of five raccoons is throwing a pool party…
Everyday realism for social
For clips that should feel filmed rather than generated, ask for imperfection — handheld framing, natural light, small mistakes — and keep the person's face, outfit and hair identical in every shot.
Prompt 15-second ultra-realistic personal home video of an ordinary rainy Sunday for a young woman in Lisbon. It should feel spontaneous, intimate and imperfect, observed rather than performed…
How it works
How to make an AI video
Pick a video model
Open the model card above the prompt box. Each model lists its tasks — text to video, image to video, references, edit or extend — and switching models keeps your prompt.
Describe the shot or add a frame
Write the prompt, or upload a first frame (and a last frame where the model takes one). Upload slots appear only for the inputs the selected model accepts.
Set length, resolution and sound
The settings show only what the model supports. The line under Generate updates the estimated credits, and the most that can be reserved, as you change them.
Generate and download
Click Generate. The clip appears in Recent and in History, where you can download it, rename it or reuse its settings for the next take.
Models
Compare the AI video models
All 32 video models with what each is good at, how long and how sharp it goes, whether it makes sound and what it accepts as input. The facts come from each model's contract and the prices from today's price list, worked out the same way the generator does.
Model list updated
Seedance 1.0 Pro Fast
BytePlusMulti-shot video from text or image
- Text to video
- Image to video
- Length
- 2–12 s
- Resolution
- 480p–1080p
- Inputs
- Up to 1 image
Gemini Omni 1.1 Flash
GoogleVideo edit and extend with 4K upscale
- Text to video
- Image to video
- Reference to video
- Edit video
- Extend video
- Length
- 3–10 s
- Resolution
- 360p–4K
- Inputs
- Up to 10 images · 3 videos
Gemini Omni Flash
GoogleKeyframe interpolation with native audio
- Text to video
- Image to video
- Reference to video
- Edit video
- Inputs
- Up to 6 images · 1 video
Grok Imagine Video 1.0
xAIText and still-image video generation
- Text to video
- Image to video
- Length
- 1–15 s
- Resolution
- 480p–720p
- Inputs
- Up to 7 images
Grok Imagine Video 1.5
xAI1080p video with lip-synced speech
- Text to video
- Image to video
- Length
- 1–15 s
- Resolution
- 480p–1080p
- Inputs
- Up to 7 images
Happy Horse 1.0
AlibabaPhysically realistic motion-smooth video
- Text to video
- Image to video
- Reference to video
- Edit video
- Sound
- Length
- 3–15 s
- Resolution
- 720p–1080p
- Inputs
- Up to 9 images · 1 video
Happy Horse 1.1
AlibabaStable subject and scene referencing
- Text to video
- Image to video
- Reference to video
- Length
- 3–15 s
- Resolution
- 720p–1080p
- Inputs
- Up to 9 images
Kling O3
KlingReference and edit video modes
- Text to video
- Image to video
- Reference to video
- Edit video
- Multi-shot
- Sound
- Length
- 3–15 s
- Resolution
- 720p–4K
- Inputs
- Up to 16 images · 1 video
Kling O1
KlingImage-to-video and video editing
- Image to video
- Edit video
- Sound
- Length
- 5–10 s
- Inputs
- Up to 4 images · 1 video
Kling 2.6 Motion Control
KlingMotion transfer from reference video
- Motion control
- Sound
- Resolution
- 720p–1080p
- Inputs
- Up to 1 image · 1 video
Kling 3.0
KlingText-to-video with element library
- Text to video
- Image to video
- Multi-shot
- Sound
- Length
- 3–15 s
- Resolution
- 720p–4K
- Inputs
- Up to 2 images
Kling 3.0 Motion Control
KlingMotion control with facial consistency
- Motion control
- Sound
- Resolution
- 720p–1080p
- Inputs
- Up to 1 image · 1 video
Kling 3.0 Turbo
KlingFaster text and image video generation
- Text to video
- Image to video
- Length
- 3–15 s
- Resolution
- 720p–1080p
- Inputs
- Up to 1 image
MiniMax H3
MiniMaxOmni-modal 2K video with stereo audio
- Text to video
- Image to video
- Reference to video
- Length
- 4–15 s
- Resolution
- 768p–2K
- Inputs
- Up to 9 images · 3 videos · 3 audio
MiniMax H3 Max
MiniMaxFast 768p video for high-volume iteration
- Text to video
- Image to video
- Length
- 5–15 s
- Resolution
- 480p–768p
- Inputs
- Up to 2 images
MiniMax H3 Max Turbo
MiniMaxText or first-to-last frame generation
- Text to video
- Image to video
- Length
- 5–15 s
- Resolution
- 480p–768p
- Inputs
- Up to 2 images
Hailuo 02
MiniMaxExtreme physics mastery for complex motion
- Text to video
- Image to video
- Length
- 6–10 s
- Resolution
- 512p–1080p
- Inputs
- Up to 2 images
Hailuo 2.3
MiniMaxExpressive characters across art styles
- Text to video
- Image to video
- Length
- 6–10 s
- Resolution
- 768p–1080p
- Inputs
- Up to 1 image
OmniHuman 1.5
BytePlusMulti-character interaction video generation
- Image to video
- Inputs
- Up to 1 image · 1 audio
Seedance 1.5 Pro
BytePlusJoint audio-video with lip-sync
- Text to video
- Image to video
- Sound
- Length
- 4–12 s
- Resolution
- 480p–1080p
- Inputs
- Up to 2 images
Seedance 2.0 Fast
BytePlusReal-person-style video for ads and UGC
- Text to video
- Image to video
- Reference to video
- Sound
- Length
- 4–15 s
- Resolution
- 480p–720p
- Inputs
- Up to 9 images · 3 videos · 3 audio
Seedance 2.0
BytePlusMultimodal references with 15s dual audio
- Text to video
- Image to video
- Reference to video
- Sound
- Length
- 4–15 s
- Resolution
- 480p–4K
- Inputs
- Up to 9 images · 3 videos · 3 audio
Seedance 2.0 Mini
BytePlusLower-cost drafts and prompt testing
- Text to video
- Image to video
- Reference to video
- Sound
- Length
- 4–15 s
- Resolution
- 480p–720p
- Inputs
- Up to 9 images · 3 videos · 3 audio
Seedance 2.5
BytePlus30-second storytelling with precise references
- Text to video
- Image to video
- Reference to video
- Edit video
- Extend video
- Sound
- Length
- 4–30 s
- Resolution
- 480p–1080p
- Inputs
- Up to 30 images · 10 videos · 10 audio
Tencent Video Upscale
Tencent720p to 4K video enhancement modes
- Upscale
- Resolution
- 720p–4K
- Inputs
- Up to 1 video
Veo 3.1 Fast
GoogleFast cinematic video with synced audio
- Text to video
- Image to video
- Sound
- Length
- 4–8 s
- Resolution
- 720p–4K
- Inputs
- Up to 3 images
Veo 3.1 Pro
GoogleScene extension and last-frame control
- Text to video
- Image to video
- Sound
- Length
- 4–8 s
- Resolution
- 720p–4K
- Inputs
- Up to 3 images
Wan 2.5
AlibabaText-to-video with synchronized audio
- Text to video
- Image to video
- Length
- 5–10 s
- Resolution
- 480p–1080p
- Inputs
- Up to 1 image
Wan 2.6
AlibabaMulti-shot narrative with audio sync
- Text to video
- Image to video
- Reference to video
- Multi-shot
- Sound
- Length
- 2–15 s
- Resolution
- 720p–1080p
- Inputs
- Up to 1 image · 3 videos · 1 audio
Wan 2.7
AlibabaMulti-shot narratives and style transfer
- Text to video
- Image to video
- Reference to video
- Edit video
- Sound
- Length
- 2–15 s
- Resolution
- 720p–1080p
- Inputs
- Up to 17 images · 1 video · 5 audio
Wan 3.0
AlibabaAll-in-one reference with docs and web
- Text to video
- Image to video
- Reference to video
- Sound
- Length
- 2–30 s
- Resolution
- 480p–1080p
- Inputs
- Up to 10 images · 5 videos · 5 audio
Wan 3.0 Prime
AlibabaFaster end-to-end all-reference generation
- Text to video
- Image to video
- Reference to video
- Sound
- Length
- 2–30 s
- Resolution
- 480p–1080p
- Inputs
- Up to 10 images · 5 videos · 5 audio
'%3e%3cmask%20id='mask0_2008_957'%20style='mask-type:alpha'%20maskUnits='userSpaceOnUse'%20x='0'%20y='0'%20width='512'%20height='512'%3e%3cpath%20d='M384%200H128C57.3075%200%200%2057.3075%200%20128V384C0%20454.692%2057.3075%20512%20128%20512H384C454.692%20512%20512%20454.692%20512%20384V128C512%2057.3075%20454.692%200%20384%200Z'%20fill='%230A0A0A'/%3e%3c/mask%3e%3cg%20mask='url(%23mask0_2008_957)'%3e%3cpath%20d='M384%200H128C57.3075%200%200%2057.3075%200%20128V384C0%20454.692%2057.3075%20512%20128%20512H384C454.692%20512%20512%20454.692%20512%20384V128C512%2057.3075%20454.692%200%20384%200Z'%20fill='white'/%3e%3cpath%20d='M838.051%2040.6242C741.469%2048.4988%20361.844%2095.5627%2098.6164%20360.182H0.935547L11.8498%20349.305C66.9346%20296.088%20310.679%2070.5472%20838.051%2040.3311V40.6242Z'%20fill='black'/%3e%3cpath%20d='M507.453%20360.182H430.893L280.046%20250.377C293.947%20241.635%20307.937%20233.309%20321.97%20225.38L507.453%20360.182Z'%20fill='black'/%3e%3cpath%20d='M385.097%20360.183H308.571L285.356%20343.298H155.747C164.872%20335.195%20174.109%20327.33%20183.444%20319.695H252.87L216.943%20293.561C229.785%20284.006%20242.769%20274.853%20255.856%20266.085L385.097%20360.183Z'%20fill='black'/%3e%3cpath%20d='M106.562%20157.711L165.573%20200.597C150.886%20209.246%20137.14%20217.805%20124.31%20226.179L30.1019%20157.675L106.562%20157.711Z'%20fill='black'/%3e%3c/g%3e%3c/g%3e%3cdefs%3e%3cclipPath%20id='clip0_2008_957'%3e%3crect%20width='512'%20height='512'%20fill='white'/%3e%3c/clipPath%3e%3c/defs%3e%3c/svg%3e){kind=small}
Pricing
How credits work for AI video
You see the cost before you generate, and you are charged for what the model actually used.
A video's price depends on the model, the clip's length, its resolution and extras such as generated sound. The line under Generate shows the estimated credits and, when it can differ, the most that will be reserved.
Clicking Generate reserves that maximum from your balance. When the clip is done you are charged for what the model reports it used, and the rest of the reservation comes straight back.
If a generation fails, its card says why and whether anything was charged. Failures on our side and generations stopped by our content check cost nothing. When the model provider fails a generation without charging for it, neither do we; when the provider charges for the compute a failed attempt used, you are charged that amount and the record says so.
Free credits, such as a sign-up bonus or daily check-in credits, work in this generator at each model's lowest quality and resolution. Higher settings and paid extras such as generated sound open up when you buy a credit pack, a one-time payment — the Pricing page lists the packs and their rules.
See credit packs and pricingFAQ
Frequently asked questions
Didn't find your answer? Email us at support@alreadyran.com.
What is an AI video generator?
An AI video generator turns a text prompt, a still image or reference footage into a new video clip. On AlreadyRan you choose one of 32 video models, describe the shot, and the model renders it; you can then download the clip or reuse its settings for the next take.
Is the AI video generator free?
Generating video uses credits. Free credits, such as a sign-up bonus or daily check-in credits, work here at each model's lowest quality and resolution. Higher settings and paid extras such as generated sound need a credit pack, bought once with no renewal; the Pricing page lists the packs.
Which AI video model is the best?
There is no single best model: they differ in motion, realism, length, resolution and sound. The comparison on this page shows what each model is good at and its limits. A practical approach is to draft short, low-resolution takes on a cheaper model, then render the final shot on the one whose look you like.
Can I turn a photo into a video?
Yes. Choose a model that lists image to video, upload your picture as the first frame and describe the motion. Some models also take a last frame and fill in the movement between the two. Starting from an image is the most reliable way to control how the clip looks.
How long can an AI video be?
Up to 30 seconds in one generation, depending on the model; most models let you choose the length before you generate. Some models can also extend a clip you already have. For longer pieces, generate the shots separately and cut them together.
What resolution are the videos?
Up to 4K, depending on the model. Higher resolutions cost more credits and take longer, so many people test at a lower resolution and render the keeper at the top setting.
Can the videos have sound?
17 of the 32 video models list a sound option, so they can generate dialogue, effects or music with the picture; turn it on before you generate. Use the With sound filter in the comparison to see which they are.
How long does a video take to generate?
From under a minute to several minutes, depending on the model, the length and the resolution. The generation keeps running if you leave the page, and the finished clip appears in Recent and in your History.
How much does an AI video cost?
It depends on the model, the length, the resolution and sound. The comparison shows each model's lowest price at today's rates, and the generator shows the estimate for your exact settings before you click. You are charged for what the model actually used.
Can I use the videos commercially?
As between you and AlreadyRan, the videos generated for you are yours to use, subject to our Terms of Use, the terms of the provider that runs the model, and the law — which in some countries does not grant copyright in machine-generated work. You are responsible for having the rights to anything you upload.
What happens to my prompts and uploads?
Your uploads are stored privately and open only for you while you are signed in. We do not use your prompts, uploads or results to train models of our own. To make a video, your prompt and reference files are sent to the provider that runs the model you chose.
What happens if a video fails or is blocked?
Every request passes a content safety check before it reaches the model. If our check or the model's own safety system stops a generation, or the model fails, the card says why and whether any credits were charged. Our check and failures on our side are never charged.
Make your first AI video
Pick a model, describe the shot and see the credits before you click Generate.