Seedance Kling Hailuo Grok Imagine

AI Video Generator

Write a shot or add pictures, pick one of 13 video models and see the price before you start. Run the same prompt on a second model to compare.

  • 13 video models
  • 3 ways to start
  • Price before every run
A 2D anime girl in a white dress and a straw hat rides a bicycle down a grassy hill toward the sea
Made with A girl in a white summer dress rides a bicycle down a hill toward the sea, the camera tracking beside her. 2D anime style. H3 Max Turbo · text to video · 5 s · 480P

1Write the shot

What to writeWho or whatWhereWhat happens

3Settings for H3 Max Turbo

Length
Quality
Format
More options
H3 Max TurboText to video · 5 s · 768P · 16:9
50 credits$0.60 on a Starter plan

2Pick a model

11 of 13 models make text to video. Prices are for 5 seconds at each model's nearest setting.

Same first frame, two models

One frame from our sample film, the camera orbiting left, 480P. The motion, the framing and the sound differ from model to model, so the quickest way to choose is to run the same shot twice.

Two swordsmen stand in tall grass at sunset while the camera orbits left, made by Grok Imagine 1.5 LiteGrok Imagine Video 1.5 Lite
Image to video · 3 s · 480P
The same scene of two swordsmen at sunset with the camera orbiting left, made by H3 Max TurboH3 Max Turbo
Image to video · 3 s · 480P

About Grok Imagine Video 1.5 Lite About H3 Max Turbo

More clips from this generator

Each one began as a single written shot. Tap a clip to hear it.

A lighthouse keeper in a yellow coat climbs a stone spiral staircase holding a lit lanternH3 Max Turbo
A lighthouse keeper in a yellow oilskin coat climbs a spiral staircase with a lantern, rain on the windows, the camera rising with her.H3 Max Turbo · text to video · 5 s · 768P
Paper lanterns rise from a dark river past a stone bridge at night, painted in watercolorH3 Max Turbo
Paper lanterns drift up from a dark river at night, a small boat under a stone bridge, the camera tilting up. Watercolor painting style.H3 Max Turbo · text to video · 5 s · 480P
A small white robot pours water on a red flower in a rooftop garden with a city glowing behind itH3 Max Turbo
A small round robot waters a single red flower in a rooftop garden at sunset, slow push in, the city skyline glowing behind it. 3D animated film style.H3 Max Turbo · text to video · 5 s · 768P

What to write for each mode

A video prompt is one shot, not a story: who or what is in it, where, what happens, and how the camera moves. Quality words such as 8k or masterpiece add nothing.

  • Text to video. Name the subject, the place, the one thing that happens and the camera: "a lighthouse keeper climbs a spiral staircase with a lantern, rain on the windows, the camera rising with her".
  • Image to video. The picture already holds the look, so write only what moves and how the camera moves: "she turns to the window and smiles, the curtain lifts in the wind, slow push in".
  • Reference to video. Name each picture in the prompt, the way the model expects it: @Image1 on Seedance and Kling, <IMAGE_0> on Grok Imagine, "Image 1" on Hailuo. Give each picture one job, a person, a place or a prop.

Weak prompt, better prompt

Weak: "a cool cinematic video of a samurai, epic, 8k". The model has to invent the place, the action, the camera and the light, so every model gives you a generic poster that barely moves.

Better: "a samurai in a gray coat walks through tall grass at dusk toward a lone tree, the camera low in the grass tracking beside him, the wind bending the grass, orange sun low behind the tree". Every part of that sentence is something a model can draw and move.

The models, one by one

Each card links to a page with what the model is, what it is good at, its limits and a price from the live quote.

ByteDance Seedance 2.5 Seedance 2.5 is ByteDance's video model for long takes. It makes clips of 4 to 30 seconds with sound from text, a first frame or up to 50 reference files, and it can also edit or extend a clip. Text, Image, Reference4 to 30 s480P to 1080P 137 credits for 5 s at 480PSee the model ByteDance Seedance 2.0 Mini Seedance 2.0 Mini is a faster, lower-cost version of ByteDance's Seedance 2.0. It makes 4 to 15 second clips with sound from text, a first and last frame, or up to 12 reference files, at 480p or 720p. Text, Image, Reference4 to 15 s480P to 720P 45 credits for 5 s at 480PSee the model Kuaishou Kling 3.0 Pro Kling 3.0 is Kuaishou's video model with sound built in. It makes 3 to 15 second clips from text or a first frame, can plan a multi-shot scene in one go, and keeps characters steady with elements. Text, Image3 to 15 s1080P 208 credits for 5 s at 1080PSee the model Kuaishou Kling 3.0 Standard Kling 3.0 is Kuaishou's video model with sound built in. It makes 3 to 15 second clips from text or a first frame, can plan a multi-shot scene in one go, and keeps characters steady with elements. Text, Image3 to 15 s720P 156 credits for 5 s at 720PSee the model Kuaishou Kling 3.0 Turbo Kling 3.0 Turbo is the faster, lower-cost Kling 3.0. It makes 3 to 15 second clips with sound from text or a first frame, holds storyboards of up to 6 shots, and has improved lip sync. Text, Image3 to 15 s720P to 1080P 139 credits for 5 s at 720PSee the model Kuaishou Kling O3 Kling O3 is Kling's omni model. It takes text, a first and last frame, reference images and character elements together, makes 3 to 15 second clips with sound, and can change an existing video from a written instruction. Text, Image, Reference3 to 15 s720P to 1080P 139 credits for 5 s at 720PSee the model Kuaishou Kling O1 Kling O1 is Kling's unified multimodal video model from December 2025. On Ziztu it makes 3 to 10 second clips without sound from a first and last frame, or from up to 7 references. Image, Reference3 to 10 s1080P 139 credits for 5 s at 1080PSee the model MiniMax H3 Max Turbo H3 Max Turbo is fal's post-trained version of MiniMax H3, tuned for prompt following and speed. It makes clips of up to 15 seconds from text or a first and last picture, at 480p, 768p or 1080p. Text, Image1 to 15 s480P to 1080P 31 credits for 5 s at 480PSee the model MiniMax H3 Max H3 Max is a faster H3 variant that MiniMax released with fal. It makes clips up to 15 seconds from text or first and last frames, from 12 reference files at most, or from a camera path over a still. Text, Image5 to 15 s480P to 768P 62 credits for 5 s at 480PSee the model MiniMax H3 Max References H3 Max is a faster H3 variant that MiniMax released with fal. It makes clips up to 15 seconds from text or first and last frames, from 12 reference files at most, or from a camera path over a still. Reference1 to 15 s480P to 1080P 62 credits for 5 s at 480PSee the model MiniMax H3 2K H3 is MiniMax's open-weights video model, the successor to Hailuo 02. It makes 5 to 15 second 2K clips with stereo sound from text, frames or references, and is strong at on-screen text and following edit instructions. Text, Image, Reference5 to 15 s2K 161 credits for 5 s at 2KSee the model xAI Grok Imagine Video 1.5 Grok Imagine Video 1.5 is xAI's video model with generated sound and lip-synced speech. It makes clips of 1 to 15 seconds from text, one picture or up to 7 reference images, at 480p, 720p or 1080p. Text, Image, Reference1 to 15 s480P to 1080P 99 credits for 5 s at 480PSee the model xAI Grok Imagine Video 1.5 Lite Grok Imagine Video 1.5 Lite is a faster, lower-cost version distilled from Grok Imagine Video 1.5. It makes 1 to 15 second clips with lip-synced speech from text or one picture, and reaches 1080p by upscaling 720p. Text, Image1 to 15 s480P to 1080P 25 credits for 5 s at 480PSee the model

What a clip costs, by model

Credits for one 5 second clip, from the same quote the Generate button uses (dollars at the Starter plan rate of $12 for 1,000 credits)
ModelMakesLengthLowest qualityHighest quality
Seedance 2.5Text, Image, Reference4 to 30 s137 credits ($1.64), 480P1,438 credits ($17.26), 1080P
Seedance 2.0 MiniText, Image, Reference4 to 15 s45 credits ($0.54), 480P96 credits ($1.15), 720P
Kling 3.0 ProText, Image3 to 15 s208 credits ($2.50), 1080POnly 1080P
Kling 3.0 StandardText, Image3 to 15 s156 credits ($1.87), 720POnly 720P
Kling 3.0 TurboText, Image3 to 15 s139 credits ($1.67), 720P174 credits ($2.09), 1080P
Kling O3Text, Image, Reference3 to 15 s139 credits ($1.67), 720P174 credits ($2.09), 1080P
Kling O1Image, Reference3 to 10 s139 credits ($1.67), 1080POnly 1080P
H3 Max TurboText, Image1 to 15 s31 credits ($0.37), 480P99 credits ($1.19), 1080P
H3 MaxText, Image5 to 15 s62 credits ($0.74), 480P99 credits ($1.19), 768P
H3 Max ReferencesReference1 to 15 s62 credits ($0.74), 480P198 credits ($2.38), 1080P
H3 2KText, Image, Reference5 to 15 s161 credits ($1.93), 2KOnly 2K
Grok Imagine Video 1.5Text, Image, Reference1 to 15 s99 credits ($1.19), 480P309 credits ($3.71), 1080P
Grok Imagine Video 1.5 LiteText, Image1 to 15 s25 credits ($0.30), 480P174 credits ($2.09), 1080P

How it works

  1. 01

    Pick what you start from

    Text to video, image to video (a first frame, and a last frame on models that take one) or reference to video with up to 9 pictures, depending on the model.

  2. 02

    Pick a model and write the shot

    Every card shows what the model makes, its longest clip, its resolutions and its price for the settings you chose. Prices come from the same quote billing uses.

  3. 03

    Generate, compare, keep

    Download the clip, use it in a project, or run the same prompt on another model and compare the two side by side.

What you get

One place for the leading video models

13 models from ByteDance, Kuaishou, MiniMax and xAI, one credit balance, no separate accounts.

The price before every run

Each card and the Generate button show the credits and the dollars for the exact settings, and a failed clip is refunded.

Built for episodes, not one-offs

Reference pictures keep a cast the same, and Ziztu Studio plans the shots of a whole micro drama or anime episode.

From one clip to an episode

One clip is one shot. A micro drama or an anime episode is a dozen of them with the same cast, and the hard part is keeping the cast the same from clip to clip. Ziztu Studio plans the shots for you, sends character sheets with every clip and lets you pick a different model for each shot, all on one credit balance. Start with the AI Micro Drama Generator or the AI Anime Video Generator, or plan the shots first with the AI Storyboard Generator.

Honest limits

  • Video starts on a plan. The free credits cover pictures, scripts and voices, not video.
  • Clips are short. The longest single clip here is 30 seconds; a longer video is several clips cut together.
  • Hands, small objects and text are the weakest parts of any AI video model, and these models are no exception.
  • Prices here are credits. The dollar figures are at the Starter plan rate, and every model page says when its facts were last checked.

Frequently asked questions

Which AI video models can I use in this generator?

13 models in four families: Seedance (Seedance 2.5, Seedance 2.0 Mini), Kling (Kling 3.0 Pro, Kling 3.0 Standard, Kling 3.0 Turbo, Kling O3, Kling O1), Hailuo (H3 Max Turbo, H3 Max, H3 Max References, H3 2K) and Grok Imagine (Grok Imagine Video 1.5, Grok Imagine Video 1.5 Lite). Each card says which modes it makes, how long a clip can be, which resolutions it renders and what it costs.

How much does an AI video cost here?

It depends on the model, the length and the resolution, and the page shows the exact price before you start. A 5 second H3 Max Turbo clip at 768P is 50 credits ($0.60), a 5 second Seedance 2.5 clip at 720P is 293 credits ($3.52) and a 5 second Kling 3.0 Pro clip is 208 credits ($2.50). Dollars are at the Starter plan rate; bigger plans and packs cost less per credit. Plans start at $12 a month for 1,000 credits.

Is the AI video generator free?

Video starts on a plan, from $12 a month for 1,000 credits. A new account gets 40 free credits, no card, for scripts, character art, storyboards, voices and sound, and the free credits pay for reference pictures from the cheaper image models here. Free results carry a small watermark. 30% off your first 3 months with the code SWITCH30, until 13 November 2026.

What is the difference between text, image and reference to video?

Text to video starts from your words alone. Image to video starts from your picture as the first frame, and on some models you can add a last frame too. Reference to video takes pictures of a person, a place or an object and keeps them in the shot while the model invents the rest. Use text for a quick look, image when the first frame matters, and reference when the same cast must appear in more than one clip.

Which model makes the longest clips, and which is the cheapest?

Seedance 2.5 makes the longest single clips here, up to 30 seconds. The lowest price for a 5 second text to video clip is Grok Imagine Video 1.5 Lite at 480P: 25 credits ($0.30). Seedance 2.5 is one of the models that take the most reference pictures, up to 9. These are numbers from the model cards, not a ranking of quality: the quickest way to choose is to run the same prompt on two models.

Can I compare two models with the same prompt?

Yes. After a clip finishes, open "Same prompt, another model" under it, pick a model and run it. The new clip appears beside the first with its model, settings and price, so you can compare the motion, the framing and the sound before you choose a model for the rest of the shots.

How do I keep the same character in every clip?

Use reference to video with pictures of your character, or start every clip from the same picture with image to video. Make the pictures with an image model in the Make pictures tab, then pick them for each clip. For a whole episode, Ziztu Studio keeps character sheets and sends them with every shot, so the cast stays the same across a scene.

Can I use the clips commercially?

Clips made with paid credits have no watermark, and every plan includes commercial use. Use pictures you have the right to use, and no real person without their permission.

Do I need an account to look around?

No. You can pick a mode, write a prompt, compare every model and see every price without an account. An account is needed to add pictures, make reference pictures and start a clip.

Related tools

A red vintage convertible on a sunny coastal cliff road above a turquoise seaText to Video AIDescribe one shot in plain words, pick a look and a quality, and MiniMax H3 Max Turbo (Hailuo 3) makes a 5 second widescreen clip in a few minutes.A photo print of the sea whose waves splash out past its edge on lemon yellowImage to Video AIUpload a photo or a drawing, say what should move, pick a camera move, and MiniMax H3 Max Turbo (Hailuo 3) makes a 5 second clip that starts on your picture.A woman in a red dress and a man in a navy suit mid-argument in a sunlit mansion hallAI Micro Drama GeneratorDescribe one scene and get a 5 second vertical drama clip with sound.An anime girl running along a seaside railway at sunset as a train passesAI Anime Video GeneratorType a scene and MiniMax H3 Max Turbo (Hailuo 3) makes a 5-second anime clip in about a minute.A six-panel storyboard sheet on a sky-blue backgroundAI Storyboard GeneratorDescribe one scene and get a six panel storyboard planned like a director would: a wide to open, an insert, a reverse angle and a close-up at the turn.Two swordsmen stand in tall grass at sunset while the camera orbits leftAI Video ModelsSeedance, Kling, Hailuo and Grok Imagine on one page each: what the model is, how long its clips run, what it is good at, its limits and what a clip costs.
AI Video Price IndexAI Video Prompt TemplateStoryboard TemplateVideo Script TemplateTikTok and Shorts Script Template
Three AI video frames on coral: a girl on a bicycle by the sea, two swordsmen at sunset, a robot in a garden

Pick a model and make your first clip.

40 free credits, no card, for pictures, scripts and voices. Video starts on a plan from $12 a month.

Try it free