FLUX 3 Max UGC Ads: Workflow, Prompts and Costs

FLUX 3 Max UGC ads are creator-style clips, a person talking to a phone camera about a product, generated in one pass with picture and sound together. On VIDEO AI ME you write the scene and the spoken line, pick 9:16, draft at 720p for 17 credits per second, then render the keepers at 1080p.

This guide is for teams that already buy UGC and want more angles per product without booking more creators. It covers what the model does and does not do for this format, a prompt you can paste, the hooks worth testing first, and what a finished ad costs once you count the drafts you throw away.

What are FLUX 3 Max UGC ads good for?

A UGC ad has one job: look like a real person found your product and had something to say about it. TikTok's own creative best practices recommend featuring people such as creators, employees or customers, and going for "a DIY or not overly polished style" so the ad sits naturally among organic posts. Three things carry that: a hook in the opening seconds, a voice that sounds like a person, and a room that sounds like a room.

FLUX 3 Max is a video model from Black Forest Labs, the team behind the FLUX image models. It is a post-trained variant of Black Forest Labs' FLUX 3 Video, which its developers describe as tuned for stronger prompt adherence and better visual quality. For UGC work, prompt adherence is the claim to check first: it means the product, the gesture and the line you wrote are more likely to be what comes back. Your first drafts will tell you whether that holds for your category.

Here is how the format maps onto what FLUX 3 Max does on VIDEO AI ME:

The last three rows matter. If your UGC process depends on a creator's recorded voice, or on holding one face steady across many clips from a set of reference photos, pick the model built for that job instead.

How do you write a prompt for FLUX 3 Max UGC ads?

FLUX 3 Max reads plain natural language. Write one paragraph per shot in a fixed order: subject and action, then camera, then light and look, then sound. Here is a complete text to video prompt for a skincare product, 9:16, 10 seconds:

A woman in her late twenties sits cross-legged on an unmade bed and holds a small amber dropper bottle close to the phone camera, then turns it so the label faces the lens. Handheld selfie framing at arm's length, eye level, slight natural shake. Soft daylight from a window on her right, no makeup, natural skin texture, phone camera look. She says, "Okay, three weeks in and my skin has never felt this calm." Quiet bedroom room tone, faint street noise through the window, the dropper bottle taps against her ring, no music.

Why each part is there:

• Subject and action is one beat: hold the bottle up, turn the label. Twenty seconds is room for one beat done well, not a montage. Three beats means three clips, joined in the editor.
• Camera uses phone language. "Handheld selfie framing at arm's length" asks for creator footage. "Cinematic dolly" asks for a commercial.
• Light and look pushes against polish. Window light, no makeup and natural skin texture are the details that make a clip read as filmed rather than produced.
• Sound comes last and is the part most people skip. The spoken line goes in double quotes with who says it, then the ambience, then the effect on the action, then music or "no music". A tidy music bed under a selfie video tells the viewer it is an ad before the line does.

Two more rules. Budget the dialogue at roughly two to three words per second, so a 10 second clip holds about 20 to 25 words; the line above is 12, which leaves room for a pause. And write every detail that matters, because the model reasons about your prompt and expands it before rendering. Whatever you leave open, it fills in for you, including the bottle colour and the room.

For more starting points by product type, see the FLUX 3 Max UGC prompt templates.

Which hooks should you test first?

TikTok's guidance says to prioritize your hook in the first 6 seconds and to introduce your proposition in the first 3. It also says it is better to use creatives with big differences when testing. For UGC that means testing hook types against each other, not five wordings of the same opener.

Write each one as its own prompt: same person, same room, same product, different first line and first action. That keeps the hook the only variable. Every claim in the line has to be one you can stand behind, because a generated creator saying it does not change who is making the claim. For the scripting side, how to write AI UGC ad scripts goes deeper on lines that sound spoken rather than written.

What do FLUX 3 Max UGC ads cost per finished ad?

FLUX 3 Max costs 17 credits per second at 720p and 29 credits per second at 1080p on VIDEO AI ME, the same for text to video and image to video. One credit is one cent of plan budget.

The per-clip price is not the number to plan around. The cost per finished ad is, because some drafts miss. Here is one product's week of UGC creative, drafted at 720p and finished at 1080p:

That is 2,138 credits, about $21.38 of plan budget, for two finished ads that each pair a hook with the demo beat in the editor, with eight drafts behind them. Put that next to what one creator video costs you today, including reshoots and usage rights, and decide from your own numbers.

On plans: the Pro plan's 5,600 monthly credits cover that cycle twice with room for extra takes, and the Starter plan's 1,400 credits cover the drafting stage for a single product. Allowances are on the pricing page, and the full arithmetic for every duration is in FLUX 3 Max pricing.

Can you start a UGC ad from a product photo?

Yes, and for physical products it is often the better route. Attach one image and FLUX 3 Max switches to image to video: your photo becomes the first frame, the prompt drives the motion, and the audio is still generated. A product-in-hand photo you shot yourself works well, because the packaging in the first frame is your real packaging rather than the model's idea of it.

The prompt changes shape. The image already carries the look, so write motion and sound, not appearance:

Her hand tilts the bottle toward the camera and taps the cap twice with one fingernail, then pulls it back. Handheld, slight push-in. She says, "This is the one that lives in my bag." Quiet cafe chatter behind her, a cup set down on a saucer, no music.

Frame the photo for where it is going. With an image attached, auto follows your image's shape and a named ratio overrides it, so shoot vertically if the ad will run at 9:16. FLUX 3 Max takes one reference image per clip on VIDEO AI ME, so each clip starts from one photo.

What should you check before a UGC ad goes live?

Every clip, before it goes into Ads Manager:

1. Listen with sound on. Did the line finish inside the clip? Is there music you did not ask for? If the generated audio is wrong for the placement, replace the track with your own audio.
2. Watch the speech and lip sync. Watch the mouth against the words at full size, not in a thumbnail.
3. Check the product. Labels, small print and logos inside generated frames are unreliable on every video model. If the brand name has to be legible, put it on screen after export, in your own editing tool.
4. Add captions in the editor. TikTok recommends captions or text overlays for context, and editor captions stay sharp and editable.
5. Settle disclosure before launch. A synthetic person giving a testimonial raises endorsement and disclosure questions. Read the AI UGC ads disclosure guide, the FTC's Endorsement Guides FAQ and the current rules of each platform you buy, and treat it as a compliance decision, not a creative one.

When the UGC cut is ready for a specific placement, FLUX 3 Max for TikTok ads covers the vertical format, sound and testing structure in more depth.

Frequently Asked Questions

Can FLUX 3 Max make UGC ads with a person talking to camera?

Yes. Write the person, the action and the spoken line into the prompt, and FLUX 3 Max generates the picture and the speech in one pass, with room tone and effects. Put the line in double quotes, say who speaks it, keep it to about two to three words per second, and watch the lip sync before the ad goes live.

How long should a UGC ad be?

There is no single correct length. TikTok's own guidance focuses on landing the hook in the first 6 seconds rather than on a fixed duration. A FLUX 3 Max generation runs 5, 8, 10, 12, 15 or 20 seconds, so a hook and one product beat fit in one clip, and longer ads join several clips in the editor.

How much does a FLUX 3 Max UGC ad cost?

A 10 second clip costs 170 credits at 720p and 290 credits at 1080p on VIDEO AI ME. A finished ad costs more than one clip because some drafts miss: five 10 second hook drafts plus two 1080p finals comes to 1,430 credits. Plan the budget per finished ad, not per generation.

Can I use my own voiceover or a creator's recording with FLUX 3 Max?

Not as an input. FLUX 3 Max on VIDEO AI ME generates its own audio every time and does not take an uploaded track or lip sync to one. You can replace the generated track with your own audio afterwards, or use Boreal, which accepts your own audio track, when the voice has to be a specific recording.

Do I need to disclose that a UGC ad is AI generated?

Disclosure rules depend on the platform, the country and what the ad claims, and testimonial-style ads carry their own endorsement rules. Read the current advertising policy of each platform you buy and the guidance from your regulator, such as the FTC in the United States, and get advice for your situation before launch.

Are AI UGC ads as effective as ads from real creators?

Nobody can answer that for your account in advance, and any blanket figure you see is someone else's result. What generated UGC changes is cost per tested angle: you can put more distinct hooks in front of your audience for the same production budget. Run them against your current creator ads and let your own numbers decide.