Built on Google Veo 3.1
AI video generator for whole films.
Most AI video generators return one clip. VeoStudio takes a script and returns a production: up to 99 shots, one cast that holds its face across all of them, dialogue and sound generated inside the picture, stitched into a single film. $25 a month, or bring your own Google key.
7-day trial · 50 tokens · no credit card
What it makes
A production, not a clip.
The interesting part of an AI video generator is not the render. It is whether shot forty still matches shot one — the same face, the same room, the same light — and whether you can fix one shot without paying for the film twice.
Per production
- Shots
- Up to 99
- Runtime
- ~13 min
- One shot
- 8s
- Long scene
- 148s · 720p
- Ratios
- 16:9 · 9:16 · 1:1
- 01
A film, from one script
Describe the film once. The AI director writes it out as structured scenes — action, camera, lighting, dialogue — up to 99 shots, and every one of them knows what came before it. Most generators hand back a clip; this hands back a production you can shoot.
- 02
A cast that survives the cut
Characters are generated once as reference stills and reused as conditioning in every shot they appear in. The person in shot forty is the person from shot one — same face, same wardrobe, same styling — without you re-describing them each time.
- 03
Sound that arrives with the picture
Veo 3.1 generates dialogue, effects and ambience inside the clip. Nothing is dubbed on afterwards, so nothing drifts out of sync when the shots are chained together.
- 04
A timeline you can re-shoot from
Every shot keeps its own mode, aspect ratio, duration and prompt. When shot twelve is wrong you re-render shot twelve for a few tokens, not the film — which is what makes the whole thing affordable in practice.
How it works
Four steps, and you see each one.
Nothing renders behind your back. The script is editable before the frames, the frames are visible before the video, and the video is billed per shot — so a film that is going wrong stops costing money at the point you notice.
The pipeline
- Script
- Gemini 3.7 Flash
- Stills
- Nano Banana
- Video
- Veo 3.1
- Music
- Lyria
- Export
- MP4
- 01
Describe the film
Gemini 3.7 Flash
A sentence is enough to start. The AI director turns it into a scene-by-scene script with camera, lighting and dialogue written per shot, and you edit any of it before a single token is spent. Script writing is free on every plan, trial included.
- 02
Cast it and draw the frames
Nano Banana
Characters and key frames are generated as stills first, from 1 to 4 tokens each depending on the image model. You see the look of a scene before you pay video rates for it, and the same stills condition every shot that follows.
- 03
Film the shots
Veo 3.1
Each generation is 8 seconds with sound. Draft at 12 tokens blocks the film out, Fast at 30 fills a timeline quickly, Quality at 80 renders the hero shots at 1080p. Mode is set per scene, so you mix them.
- 04
Stitch and export
Movie maker
Shots drop onto one timeline, music goes over the top, and the film exports as MP4 in 16:9, 9:16 or 1:1. Commercial rights are included on every plan, with no export fees and no watermark.
The pipeline is written out shot by shot on how VeoStudio works, and the editor itself is covered on the features page.
Features
What you actually get.
Eight things that matter once you are past the first clip and trying to finish something.
Engine
- Video
- Veo 3.1
- Stills
- Nano Banana
- Director
- Gemini 3.7 Flash
- Music
- Lyria
- Also
- Runway · Kling · MiniMax
Multi-scene productions
Up to 99 shots in one production, written from one script and cut into one film — roughly 13 minutes of finished runtime.
99 shotsCharacter consistency
Reference stills per character, stored with the production and injected into every shot. Upload your own references if the cast already exists.
1–4 tokens per stillScene extension
Last frame of one shot becomes the first frame of the next, carrying a single continuous scene to 148 seconds at 720p.
Up to 148sNative audio
Dialogue, sound effects and ambience generated inside each clip by Veo 3.1. Lyria writes the music over the finished timeline.
In-clip, per shotThree render tiers
Draft to block out, Fast to fill a timeline in parallel, Quality at 1080p for the shots that carry the film. Set per scene, not per project.
12–80 tokens per shotVertical and square
16:9, 9:16 and 1:1 are all native and set per production, and the ratio carries through to the stitched export for TikTok, Reels and Shorts.
16:9 · 9:16 · 1:1Bring your own key
On Pro, plug in a Google AI Studio key and rendering runs on your own Google account at Google's prices — uncapped, and zero tokens here.
Pro featureCommercial rights included
Client work, resale, ads, festival submissions. Every plan including the trial carries full commercial licensing, with no per-export fee and no attribution requirement.
All plansThe limits
What no AI video generator can do yet.
These are constraints of the model, not of the plan you buy. Anyone selling around them is selling clips glued end to end, so it is worth knowing the shape of them before you pay anybody, including us.
- 8s per render
- Veo 3.1 accepts 4, 6 or 8 seconds per generation, on every plan and in every product built on it. Length comes from chaining shots, never from a longer single render.
- 148s means 720p
- Scene extension runs at 720p only, so a 148-second continuous scene and 1080p are mutually exclusive. Full HD is Quality mode’s native 8-second shot, which you cut between rather than extend.
Fast mode renders shots in parallel to save time, which means it has no previous frame to hand forward — chaining runs on Draft and Quality only. The whole length question is worked through on AI video longer than 8 seconds.
What it costs
$25 a month, and you can see every charge.
Rendering is billed in tokens, per shot and per still, so a film costs what it renders rather than what tier you sit on. Writing the script and directing the production are free on every plan, trial included.
| What | Resolution | Available on | Tokens |
|---|---|---|---|
| Draft shot (8s) | 720p | Every plan | 12 |
| Fast shot (8s) | 720p | Every plan | 30 |
| Quality shot (8s) | 1080p | Pro | 80 |
| Still image or frame | — | Every plan | 1–4 |
| AI script writing | — | Every plan | Free |
In plain terms: the 7-day trial’s 50 tokens film about 4 eight-second Draft shots, enough to write a production, cast it and see whether the continuity holds. Pro is $25 a month for 300 tokens — roughly 25 Draft shots — and it opens Quality mode and the option to run everything on your own Google key instead. The full ladder, including token packs, is on the pricing page.
Questions
Straight answers.
What is an AI video generator?
Software that turns a written description into finished footage. You describe a shot; a video model renders it, in VeoStudio's case Google Veo 3.1, with the dialogue, sound effects and room tone generated inside the same clip. What separates one generator from another is mostly what happens around that render — whether you get a single clip back, or a script, a cast and a timeline that hold together across a whole film.
Can an AI video generator make more than one scene?
This one does. A VeoStudio production holds up to 99 shots written from a single script, with the same characters and the same look running through all of them. A single Veo generation is still 8 seconds; the film is those shots chained and stitched, which is a different thing from generating 99 unrelated clips and hoping they cut together.
How long can an AI-generated video be?
One generation is 8 seconds on Veo 3.1, and no plan anywhere lifts that. Continuity frames chain shots into one continuous scene of up to 148 seconds, and a production of 99 shots stitches to roughly 13 minutes. The chained long scene renders at 720p — 148 seconds and 1080p are not available together.
Does the AI video have sound?
Yes. Veo 3.1 generates dialogue, sound effects and ambience inside the clip rather than laying them over the top afterwards, so lips, footsteps and room tone stay in sync when shots are chained. Music is added to the finished timeline separately, generated with Lyria.
Do the characters stay the same across scenes?
Yes, and it is the reason to write the script first. A character is generated once as reference stills, stored with the production and injected into every frame and every shot that features them, so face, wardrobe and styling hold from shot one to shot ninety-nine. You can upload your own reference images instead.
What does it cost?
The trial is 7 days, 50 tokens and no credit card — about 4 eight-second Draft shots. Pro is $25 a month for 300 tokens, roughly 25 Draft shots, and it opens Quality mode at 1080p. Video runs 12–80 tokens per shot depending on the tier; script writing is free on every plan.
Can I use my own Google API key?
Yes, on Pro. Add a Google AI Studio key and video, images, frames and music generate on your own Google account at Google's prices, costing zero tokens here. Script writing and the AI director are not on your key — they run on the studio account and are free either way.
Start
Write something today and film it.
The 7-day trial grants 50 tokens and asks for no card — about 4 eight-second Draft shots, with the script writing, the casting and the frames on top of that for free.
7-day trial · 50 tokens · no credit card