Vizard Agent

How to make a film with no logos, branding or text anywhere in it

Last updated 2026-08-17 · 6 min read

Paste your brief into Vizard Agent and state the constraint plainly: no logos, emblems, branding or text on clothing or equipment anywhere. Vizard Agent studies your previous delivery for its look, generates the narration and every scene against that restriction, and checks the frames rather than trusting the instruction held.

What is the short version?

Sometimes a film has to be free of branding entirely — for licensing, for a partner's rules, or because the video will be shown by people who cannot be seen endorsing anything. That is a constraint the whole generation has to respect, not a cleanup pass afterwards.

  1. Go to Vizard Agent and paste the brief.
  2. State the no-branding rule explicitly, including clothing and equipment.
  3. Say the length, the shape and the register.

What do you need before you start?

The brief and the rule, written out. Vizard Agent applies a stated constraint to every generation rather than to the finished film, which is the only way it actually holds — a mark generated into a shot cannot be removed cleanly afterwards.

What do you type into Vizard Agent?

Put the constraint at the top, before the creative brief. Vizard Agent treats a stated restriction as something to apply to every scene it generates, and stating it once at the start is far more reliable than adding it to each scene description.

Prompt

Strict instruction: no logos, emblems, branding or text on clothing or equipment anywhere in this video. [Then the brief: the scenes, the narration, the length and the style.]

Variants worth knowing:

What does Vizard Agent actually do?

Here is the order Vizard Agent worked in on a real unbranded film. It starts by studying what it delivered for you before — including transcribing its own earlier narration — so the new film sounds and looks like a continuation.

  1. Checks the workspace and reads the project ledger for past work and uploads.
  2. Probes the previously delivered film for its resolution, frame rate and duration.
  3. Extracts keyframes from it and views them tiled to analyse the grading and style.
  4. Runs transcription on the earlier film to see how its narration was written.
  5. Reads the earlier uploads to check whether any were style references or logos.
  6. Loads the film playbook and reads the options for the generation, voice and music tools.
  7. Searches the music library four different ways and downloads three candidate tracks.
  8. Generates one voiceover line first to test the tone before committing.
  9. Generates the remaining lines concurrently in batches, then builds the scenes against the no-branding constraint.

Step 8 is the cheap check that saves the run. One line tells you whether the voice is right; ten lines in the wrong voice is ten regenerations, and Vizard Agent spent a few seconds finding out first.

What does the result look like?

From the run this page is written from, probed on the delivered file: 1280x720, H.264, 24fps, 120.41 seconds, AAC audio. Wide, two minutes, generated scenes with narration and a cinematic score, and no visible branding on clothing, equipment or in the background.

Vizard Agent studied the previous delivery's transcript as well as its frames, so the new film's writing sits in the same register as the one before it rather than starting from a blank page.

Two minutes across a set of generated scenes means each one runs eight to twelve seconds. Vizard Agent holds them longer than a social cut would, because this format is shown in a room rather than scrolled past on a phone.

This category was not separately measured, so treat the timing as a range rather than a promise: comparable work runs a median of 28 to 38 minutes end to end. Across all projects the median cost by tier is Flash 47, Pro 55, Max 242, Ultra 263 credits.

When does this not work well?

A negative constraint is much harder to guarantee than a positive one. Vizard Agent applies the rule during generation rather than as a cleanup afterwards and checks the frames it produced, and generated imagery still has a persistent habit of inventing plausible-looking text on signs, shirts and equipment.

How do you fix a result that came back wrong?

Name the shot that broke the rule. Vizard Agent keeps every generated scene, each narration line and all three music candidates as separate assets, so regenerating a single shot leaves the rest of the film, its narration and its timing entirely untouched.

How does Vizard Agent compare to doing it yourself?

By hand this is a shoot with a wardrobe rule, tape over every logo on the equipment, and a post pass removing the ones somebody missed — which is precisely why unbranded films are expensive to make properly.

A shoot Vizard Agent
Enforcing the rule Tape and a continuity check Applied to every generation
A mark that slipped through Paint it out in post Regenerate that shot
Matching an earlier film Reference and hope Frames and transcript studied
The narration A booth and a session Tested on one line, then generated

Common questions

Why would a film need no branding? Licensing, partner rules, or an audience that cannot be seen endorsing anything. It is a common institutional requirement.

Will it really keep every logo out? Vizard Agent applies the constraint to every generation and checks the frames. Generated signage is the case to watch.

Can I make one exception? Yes. Say which — a final card, for instance — and Vizard Agent keeps the rest clean.

Can it match a film you made for me before? Yes. Vizard Agent studies the earlier delivery's frames and its transcript before writing anything new.

How long can it be? Two minutes is comfortable. Cost scales with the number of scenes rather than the runtime.

Can it avoid on-screen captions too? Yes. Say so and Vizard Agent delivers the film with narration and no text on screen at all.

Does it check the rule was kept? Yes. Vizard Agent reviews the generated frames rather than assuming the instruction held throughout.