Vidu Image Generation and Editing
Stills and clips run on the same models in the same workspace — which removes the step where an approved image gets rebuilt somewhere else.
The Vidu models available in OpenFilmo are not video-only. The same generator panel produces still images, so a product shot, a character sheet and a clip of that character can all come out of one workspace with one credit balance.
That matters more than it sounds. The usual image-to-video workflow involves generating a still somewhere, approving it, exporting it, and then uploading it again to animate it — and each hop is a place where resolution or colour drifts. Keeping both steps in one place removes the hops.
What the image side does
The same two models, applied to stills.
Text to image
Describe a scene and generate a still, with framing chosen from the same crop set used for video so a later animation keeps the composition.
Reference to image
Feed in a subject and generate new images of it — the still-image counterpart to the consistency approach used across shots in video.
Local edits and material changes
Adjust part of an image rather than regenerating all of it, which is the difference between a usable iteration loop and starting over each time.
Straight into video
An approved still is already in the workspace, so animating it is the next action rather than a separate upload to a separate service.
How a still becomes a clip
One workspace, no export step in the middle.
Generate the still
Work at the framing you intend to animate and at the resolution the final delivery needs. It is easier to decide composition now than to correct it after motion is in.
Fix locally instead of regenerating
Correct the part that is wrong. A material swap or a local edit keeps the rest of the frame intact, which protects any consistency you have already established.
Approve before animating
Motion amplifies whatever is already in the still, including its mistakes. Check edges, hands and any text before spending credits on a clip.
Animate, then upscale if needed
Animate the approved frame, and reserve upscaling for the final export — upscaling earlier enlarges artefacts instead of removing them.
Where this gets used
Cases where holding one workspace actually shortens the job.
Product imagery
Generate the shot, adjust the detail, then animate the same asset for a listing video without a reshoot or a second tool.
Character sheets
Build a consistent set of angles for a character and use those as the references that keep later clips on-model.
Ad creative variants
Produce a set of framings for one campaign where the subject is recognisably the same across every placement.
Thumbnails and covers
Where the still is the deliverable and the point is a clear subject at small sizes rather than detail that disappears.
Why one workspace beats two tools
Every hand-off between tools is a place where quality quietly drops.
The conventional image-to-video path exports a still, uploads it elsewhere and hopes the second service interprets it the same way. Resolution gets resampled, colour shifts, and the thing you approved is no longer quite the thing you are animating.
Keeping generation and animation on the same models in the same balance removes that class of problem, and it also makes cost legible: an image is two credits and a five-second clip is five, so a test cycle has a price you can predict before you start.
The honest caveat is that stills are not where these models are most impressive. Text inside an image remains unreliable, so anything with more than a few words needs proofreading, and anything where a wrong number would be expensive should not be generated at all.
Credits and plans
Generation is metered in credits, shared across every tool.
One image costs 2 credits and a 4K upscale costs 4. The Free plan gives you 30 credits every month with no credit card, no watermark and commercial use included. Paid plans start at $9 for 400 credits and add faster generation and parallel tasks.
Related guides
Longer reads on the same workflow, from the OpenFilmo blog.
AI Video Prompt Formulas That Hold Up in Production
Prompt structures that work for AI video: subject, motion, camera, time and style — with tested formulas, failure patterns and a m…
Free AI Video Generator: No Watermark, No Card
Compare free AI video generators in 2026: free tiers, watermarks, resolution caps and commercial rights — plus when upgrading actu…
Text in AI Video: Why It Breaks and What to Do Instead
Why AI video models garble text and logos, what actually works instead, and a production workflow for adding typography to generat…
Related workflows
Closely related tools and guides on the same problem.
Frequently asked questions
Are the image models the same as the video models?
Yes — the same Vidu models appear in both the image and video sides of the generator panel, which is why an approved still animates without changing tools.
What does one image cost?
Two credits. The Free plan includes 30 credits a month with no credit card, which is roughly fifteen images, and paid plans start at $9 for 400 credits.
Can I edit an image instead of regenerating it?
Yes. Local edits and material changes let you correct one part of a frame while leaving the rest untouched, which is usually faster and cheaper than another full generation.
How is resolution handled?
Choose the output you need at generation time, or upscale afterwards for four credits. Upscaling after the final version is approved avoids spending credits on renders you will discard.
Does it handle text in images well?
Not reliably, and no current model does. Short phrases usually survive; longer lines and anything numeric should be treated as something to proofread or to add outside the image.