Job-cost screens are Qwen slides, not Veo math

Build the slide in Qwen. Turn the slideshow into a video. Veo will invent numbers.

A job-cost screen is labor hours, burden, markup, a phase code a superintendent will check against the estimate. Those strings are the product. A cinematic take of a tablet on a trailer desk with plausible digits is a cousin budget. Letter the board. Then convert the board. Do not ask a video model to do the arithmetic.

The screen is the product

Procurement pauses. Superintendents pause. They read. They do not need a talking estimator walking "what you should bid." They need a board that survives a freeze-frame: the labor line, the material line, the contingency line, the exact strings from the estimate you will actually send. Put that board on a slide. Proof it at 100%. Do not bury it in a video prompt and hope the glyphs hold for six seconds.

Qwen Image 3 Text to Image is the stills row when the letters have to read: third-generation Qwen, strong prompt adherence, complex text in Chinese and English, prompt rewriting on by default, 4 credits, 2K max, qualities 1K and 2K. No audio. No duration.

Write the exact strings. If a bid line must stay verbatim, say so in the prompt and inspect the still. Rewriting is a help for scene language, not a licence to paraphrase a burden factor. Set 1K while the words are in draft. Set 2K when the slide is the slide. There is no 4K chip on this row. A video model is a bad spreadsheet. It will give you almost-hours and a tablet legal cannot ship. Lock the characters here. Then consider motion.

Qwen letters the estimate. Veo invents the math

Do not ask Veo 3.1 — native audio, official 4K tier, 4 / 6 / 8 second clips — to "show a job-cost screen with the numbers." It will invent motion and invent lettering. Pause any such clip: the hours are the model's, not yours.

A constructed estimator walking markup or "here is what you should carry" is finance-adjacent advice in a hard-hat. YouTube's third inauthentic-content bucket is AI-generated personas that present as human experts on health, legal issues, finances, or politics. A synthetic face prescribing a bid is the same shape as a synthetic host prescribing a 401(k). Keep lips off the spec. The board is the expert.

The job-site photograph is a different object. If you even need motion, that is image-to-video from an approved still of this tablet. Prompt only camera. Do not re-describe the hours in the video paragraph. The slide already holds the math.

Plan the slideshow. Convert after the board is locked

Generate a slideshow from a prompt is the named agent job when you need a planned set: theme, format, slide count, then a board per beat. Cost is priced per image generated, shown before you confirm. That planner is useful for structure — labor, then materials, then markup, then CTA. It is dangerous if you let it invent the numbers.

Do not brief "typical job-cost screen, commercial remodel." Brief the actual strings, slide by slide, or generate the lettered stills on Qwen and assemble them as the set. A prompt-slideshow that fills hours because they sounded construction-shaped is a cousin bid. Slide count is a width decision. Do not ask for twenty tiles to "see options."

Turn a slideshow into a video is convert_slideshow_to_video: it renders a slideshow you already have. It does not write slides. duration_per_image is one hold for the set. use_edited_images is whether stamped overlays come along. Forget that flag and you render bare originals. Fix, replace, or reorder on the board first. A convert of a draft is a throwaway MP4.

Stamp a line that may change — a campaign CTA, a date — after the pictures are keepers. Caption-in-image paints letters into pixels you cannot edit. If the burden will change in legal review, stamp, do not bake.

Contractors get a different /for. This is the screen

Contractors already have an audience page: process transparency, demo-day time-lapse, a problem-and-fix on site, 9:16, two posts a week, Facebook and Reels. That page is the wall you have not closed yet. It is not a job-cost UI. A general contractor can film framing. This listing has to letter this estimate. Project-manager async standups are a third object. Status talk does not typeset burden.

Do not rewrite those pages. Photograph the screen only if the UI is already true in the pixels. Otherwise Qwen the slide, proof the characters, convert the signed set. Keep a mouth off the generate.

The test: cover the logo. Can a superintendent still read the hours? If the answer is a pretty trailer and a talking hat, you made a contractor reel with invented math. If the answer is your board, paused, you made the screen.

FAQ

Can I put the hours in a Veo prompt so the tablet typesets itself?

No. Veo will invent plate math. Letter the spec on Qwen Image 3 at 4 credits, 2K, and proof the characters. Convert the signed slideshow. Motion of a trailer, if any, is a later still-to-motion pass of a photo you already own.

Should the prompt-slideshow invent a typical burden if I only said "job cost"?

No. The planner will fill something that looks like an estimate. That something is not your bid. Write the strings, or generate Qwen stills that already contain them, then assemble.

When do I stamp overlay instead of baking type into the Qwen still?

Stamp when the line may change — CTA, campaign, a legal rewrite. Bake only when the type is the scene and you will not edit it. Do not chain a stamp on top of baked-in type or you double-print the hours.

Is this the same as the contractors audience page?

No. Contractors is the crew and the wall. This page is the job-cost screen as a lettered slide. Do not rewrite that /for. Do not glue a talking estimator onto either file.