Take a Paragraph to a Movie You Can Send
A finished piece is a chain of stages, and every one of them has its own way of failing.
Text to movie is the more literal of the two phrases: not a shot that merely looks cinematic, but a paragraph turned into a piece that can be handed to a client, published to a channel or submitted against a deadline. That chain has stages, and none of them disappears because generation improved. The text is broken into beats, each beat becomes its own generation, the takes are checked against each other, dialogue and music and captions are laid on, the whole thing is cut, and only then does an export setting matter. The workspace is OpenFilmo, running in the browser at app.openfilmo.com with nothing to install; text-to-video begins in AI Create, and Vidu Q1 and Vidu Q2 are selected per generation in the generator panel.
This page follows that chain end to end and is plain about which parts remain manual: nothing here turns one sentence into a finished movie, and the assembly is yours. What is deliberately left out is how to make any single shot look like cinema — lens choice, camera movement, light, colour and pacing belong to the text to film page. Go there if your beat list is settled and the shots simply do not feel like film; stay here if they feel right and the problem is that they are not yet a movie.
What it does
Built for this specific job, not a generic editor.
Beat list before any generation
A paragraph is not a shot list. Break it into beats of a few seconds, decide what changes inside each one, and write camera and light once per beat rather than once for the whole piece. Projects that fail usually fail here, before a single credit is spent.
One reference set for the whole run
Whatever has to look identical across shots — a person, a product, a room — should be described or supplied identically every time. Stills and clips share the same models and the same rate here, so building an image first and reusing it costs two credits and saves a reshoot.
Voice, music and captions as separate passes
Dialogue, score and subtitle work happen against picture, not inside it, and each has its own page in this workspace. Keeping them as separate revisions is the point: a caption track you can still edit is worth more than words baked into pixels.
Assembly, rhythm and export settings
The cut is where pacing is actually decided, since every shot arrives already fixed at a few seconds. Export is the last decision rather than the first — choose the resolution and aspect the destination asks for, and do resolution work once on the finished cut instead of per take.
How it works
Four steps, no timeline to learn.
Mark where each shot ends
Read the paragraph and put a mark at every point where a camera would cut. A beat is roughly one five-second generation, so a thirty-second piece wants six beats before you allow for any retakes at all.
Lock the subjects before shot one
Decide now what has to match across the cut and fix it: the same reference images, the same wording for appearance, the same lighting direction. Changing any of those halfway through the run is what produces an edit that jumps.
Approve shot by shot, in order
Work down the list keeping the takes that match rather than the ones that impress. A shot that breaks continuity gets regenerated immediately, while the references are still in front of you, not after the rest are finished.
Lay sound, cut, then export once
Put dialogue and music against picture, add and proofread captions, cut to the rhythm the piece needs, then export at the specification the destination wants. One export at the end, rather than one per clip.
What people use it for
The jobs this actually gets used for.
A thirty-second spot from a brief
Six beats, a hook inside the first, a product or presenter that must look identical in all of them, and a caption track that still carries the offer when the sound is off.
A teaser or trailer cut
Short beats, hard cuts, music doing most of the work. Trailers forgive a lot of continuity trouble because no shot stays on screen long enough to be examined, which makes them the easiest first delivery.
A dialogue-led short
Scenes built from lines and reactions, where the real risk is a face changing between cuts. Generate more takes than you need on the character beats and approve those last.
One cut for several placements
A single assembly delivered in the framings and lengths each channel asks for: generated once, exported per placement, rather than re-shot for every destination.
Why a movie is a pipeline problem, not a prompt problem
The prompt decides how a shot looks. The pipeline decides whether the piece survives assembly.
Each stage has its own failure mode and they do not fail at the same moment. A beat list that skips a transition leaves a hole you only notice in the edit. References that drift leave a cut where the lead changes face at the halfway point. Sound laid before the picture is locked has to be done again. None of those are prompt problems, which is why better phrasing rarely rescues a piece that is not working.
The arithmetic also forces the pipeline to be planned rather than improvised. At five credits per five-second clip, a thirty-second piece is six beats minimum and nearer eighteen generations once retakes are counted — 90 credits, comfortably inside the 400 that Starter at $9 provides, and far outside the 30 that Free gives each month. Knowing that number before starting is the difference between a finished piece and eleven good shots.
The honest limits sit in the assembly rather than in the generation. Nothing here edits the cut for you, matches one shot’s colour to its neighbour, or produces captions you can ship unread. Dialogue and score are added in their own passes, and the export is a setting you choose. What the pipeline buys is not automation but visibility: when the piece fails, it fails somewhere you can locate and fix instead of somewhere inside a single generation.
Credits and plans
Every tool draws on one shared credit balance.
Charging is per unit rather than per project: five credits for a five-second clip and two for an image, both from one balance, with image-to-video and text-to-video drawing on the same models at the same rate. Free is $0 a month for 30 credits and no credit card — six five-second clips, or fifteen images — with one task at a time, standard speed, 720p export, no watermark and commercial use included. Starter at $9 raises that to 400 credits, about eighty five-second clips, with two parallel tasks and every model available; Pro at $19 gives 1,000 credits, three tasks and a priority queue; Studio at $59 gives 4,000 credits, six tasks and API access. Prices are those listed on the pricing page in October 2026. Multiply your beat count by three when planning, because retakes are the largest line in any real budget.
Related guides
Longer reads on the same workflow, from the OpenFilmo blog.
Character Consistency in AI Video: What Actually Works
How to keep characters consistent across AI video shots: reference images, prompt discipline, key frame locking and post-hoc corre…
AI Short Drama Production: Costs, Workflow and Reality
How AI short drama is actually produced: real cost structure, episode workflow, where consistency breaks, and why distribution is …
AI Video Cost Per Minute: The Real Numbers in 2026
How to calculate true AI video cost per finished minute: generation price, regeneration waste, edit time and iteration, with worke…
Related tools
Closely related pages on the same problem.
Frequently asked questions
Can one prompt really produce a whole movie?
No. One generation produces one shot of a few seconds, and a movie is a list of such shots cut together. The value here is that each stage is visible and cheap to iterate, not that the assembly happens on its own.
What does a short piece cost in credits?
Five credits per five-second clip. A thirty-second piece is six beats at minimum and realistically eighteen generations with retakes, which is 90 credits — inside Starter at $9 for 400 credits, and well outside the 30 Free gives each month. Images are two credits each when you want a still to reuse as a reference.
How long can a single generation be?
The unit quoted on the pricing page is a five-second clip, and no maximum length is published. Working one beat per generation and assembling afterwards is reliable; trying to stretch a single take across a whole scene is not.
Can I use the finished piece commercially?
Yes. Commercial use is included on Free as well as on paid plans and exports carry no watermark. You still need the rights to any script, music or reference images you supply — processing material does not transfer ownership of it.
How is this different from text to film?
This page is the delivery chain: beat list, per-shot generation, dialogue, music, captions, cut and export. The text to film page is about making one shot read as cinema — lens, camera movement, light, colour and pacing. Start there if your shots do not feel like film yet; stay here if they do and the piece is still not finished.
Halloween pages
Scene ideas, prompt libraries and a step-by-step walkthrough for the Halloween effect.