How to Use FLUX 3 Max: Step by Step for Video Ads

Here is how to use FLUX 3 Max: open the VIDEO AI ME editor, choose FLUX 3 Max in the video model picker, attach nothing for text to video or one image for image to video, set duration, resolution and aspect ratio, then write one paragraph covering subject, camera, light and sound. Draft at 720p, deliver at 1080p.

That is the whole loop. The rest of this guide is the detail that decides whether your first clips are usable: which settings to pick for which placement, how to write a prompt the model follows, what to check before you spend more, and how to finish the clip in the editor. It is written for people producing ads, so every step is tied to cost per usable asset.

What do you need before you use FLUX 3 Max?

Three things, and none of them is technical:

• One job for the clip. A hook for a 9:16 ad, a product beat for a landing page, a 21:9 opener for a brand film. One scene, one idea.
• The line, if there is one. Any spoken words, written out, with who says them.
• An image, if you have the look already. A product photo or campaign still you want as the first frame.

Generation runs on your plan's credits. FLUX 3 Max costs 17 credits per second at 720p and 29 credits per second at 1080p, and text to video and image to video cost the same. A 10 second draft at 720p is 170 credits.

Step 1: Decide the shot before you open the editor

A vague brief wastes generations, because the model fills every gap you leave. Before you touch a setting, answer four questions in a sentence each: who or what is on screen, what single thing happens, where the ad runs, and what the viewer hears.

Then pick a length that fits one beat. Twenty seconds is the ceiling and every length up to it is one generation, but twenty seconds is room for one beat done well, not a montage. If your idea has three scenes, plan three clips and join them in the editor later.

Step 2: Open the editor and pick FLUX 3 Max

In the VIDEO AI ME editor, open the video model picker and choose FLUX 3 Max. It sits alongside the other models in the catalogue, including Seedance 2.5, Kling, Happy Horse, MiniMax H3 Max and Boreal, so you can switch model from one clip to the next.

If you are new to the model, the explainer on what FLUX 3 Max is covers where it comes from and what its developers claim for it.

Step 3: Choose text to video or image to video

You do not pick the mode from a menu. It follows from what you attach:

Use text to video when you have an idea and no footage. Use image to video when the look already exists, such as a product shot you have paid for. FLUX 3 Max takes one reference image on VIDEO AI ME. If your brief needs several, Seedance 2.5 and MiniMax H3 Max take several references.

The full image route is in the FLUX 3 Max image to video guide.

Step 4: Set duration, resolution and aspect ratio

The editor offers:

Resolution. Draft at 720p. You are testing the prompt, not the pixels, and a 10 second draft costs 170 credits at 720p against 290 at 1080p.

Aspect ratio. auto renders 16:9 for text to video and follows your image for image to video. A named ratio overrides that. Match the ratio to where the clip runs:

TikTok lists its accepted ratios in its video ad specifications, and YouTube notes in its upload settings help page that 16:9 is the standard ratio on a computer. Check the spec for every placement you buy before you generate. Every ratio and length is covered in the durations and aspect ratios guide.

Step 5: Write the FLUX 3 Max prompt in four parts

FLUX 3 Max reads plain natural language. Write one paragraph per shot, always in this order:

1. Subject and action. Who or what, doing what. One focused beat.
2. Camera. Framing and movement: "slow push-in", "handheld, eye level", "locked-off wide".
3. Light and look. Time of day, light source, grade.
4. Sound. Spoken lines in double quotes with who says them, then ambience and effects, then music or "no music".

A text to video example, 9:16, 10 seconds:

A woman in her thirties holds a glass bottle of cold brew up to the camera in
a sunlit kitchen, then takes a sip and smiles. Handheld, eye level, slight
push-in. Warm morning light from a window on the left, soft natural grade.
She says, "This is the only coffee I make at home now." Quiet kitchen
ambience, the clink of the bottle on the counter, no music.

An image to video example, where the image already carries the look, so the prompt is about motion and sound:

The bottle slowly rotates on the marble counter as condensation runs down the
glass. The camera makes a slow half-orbit. A crisp cap twist, a soft fizz,
light lo-fi music.

Rules that keep the model on your brief:

• The model expands short prompts. It reasons about your prompt and fills in whatever you leave open. If the product colour, the line or the framing matters, write it.
• Budget the dialogue. Roughly two to three spoken words per second, so a 10 second clip holds about 20 to 25 words. Write fewer than you think.
• Name the sound on the action. "The pump clicks as her thumb presses down" lands better than "sound effects".
• Write the ambience even when it feels obvious, or the model decides for you.
• Keep on-screen text out of the prompt. Add captions in the editor and other text after export.

Deeper prompt craft lives in the FLUX 3 Max prompt guide.

Step 6: Generate at 720p and review with sound on

Run the draft and watch it where the ad will run, on a phone for a 9:16 clip, with sound on. Check five things:

Always watch generated speech and lip sync before a clip goes live. Small on-screen text and exact logos inside generated frames are unreliable on every video model, which is one more reason to keep them out of the prompt.

Step 7: Change one thing per retry

When a draft misses, change one variable and generate again. If the action is wrong, rewrite the action sentence and leave the camera alone. If the line clips, cut words before you change anything else. If the sound is generic, name the specific sound on the action.

Step 8: Render the winner at 1080p and finish in the editor

Once a 720p draft is right, rerun the same prompt at 1080p. The clip lands as an MP4 in your project, ready for the editor, captions and export.

Finish there:

• Captions and text. Captions added in the editor and other text added after export stay sharp and editable.
• Logo. Added after export, in your own editing tool, never generated inside the frame.
• Audio. Audio is always generated with FLUX 3 Max. If you want a different track, replace it with your own audio.
• Sequence. Join several clips into one ad if your concept has more than one beat.

A typical session costs less than people expect. Four 10 second drafts at 720p are 680 credits, and one 10 second final at 1080p is 290 credits, so 970 credits gets you from idea to finished clip. For what that means against each plan, see the FLUX 3 Max pricing guide or the pricing page.

Frequently Asked Questions

How do I access FLUX 3 Max?

Open the VIDEO AI ME editor and choose FLUX 3 Max in the video model picker. Generation runs on your plan's credits, at 17 credits per second at 720p and 29 credits per second at 1080p. Agents and developers can also reach it through the make_video skill on the VIDEO AI ME MCP server and /api/v1.

How do you write a prompt for FLUX 3 Max?

Write one paragraph per shot in four parts: subject and action, then camera, then light and look, then sound. Put spoken lines in double quotes with who says them, then ambience and effects, then music or no music. The model expands short prompts, so write any detail that matters.

Can I use my own image with FLUX 3 Max?

Yes. Attach one reference image and FLUX 3 Max uses it as the first frame, with your prompt driving the motion and the sound. Because the image already carries the look, describe motion and sound rather than appearance. It takes one image only on VIDEO AI ME.

What aspect ratio should I use for FLUX 3 Max?

Match the placement. Use 9:16 for vertical short-form feeds, 16:9 for YouTube and landing pages, 1:1 for square feeds and 21:9 for cinematic brand work. The editor also offers 4:3, 3:4 and auto, which renders 16:9 for text to video and follows your image for image to video.

How much dialogue should a FLUX 3 Max prompt include?

Roughly two to three spoken words per second of clip, so a 10 second clip holds about 20 to 25 words and a 5 second hook about 10 to 15. Write fewer than you think, because a line that runs long can sound rushed or get cut off.

Can you replace the audio in a FLUX 3 Max video?

Yes, in the editor. FLUX 3 Max always generates sound in the same pass as the picture on VIDEO AI ME, so every clip arrives with audio. If you want a different track, replace the generated audio with your own audio after the clip lands.