FLUX 3 Max Text to Video

Image Gen·flux-3·by Black Forest Labs

FLUX 3 Max Text-to-Video turns text prompts into 720p or 1080p videos with optional audio. Create 5–20-second clips for stories, concepts, and social media.

Runtime (p50)
-
Estimated price
$0.0425–$0.0725 per second, depending on resolution
Call the API
prediction.sh
sh
curl -X POST \
  -H "Authorization: Bearer $EACHLABS_API_KEY" \
  -H "Content-Type: application/json" \
  --data '{
    "model": "flux-3-max-text-to-video",
    "input": {
        "prompt": "Cinematic haute couture runway film. Locked-off frontal camera at the end of the catwalk with a very slow push-in. A monumental dark stone hall with tall fluted columns along both walls, a polished dark stone floor, small white cube lights spaced along both sides of the runway, and a single tall vertical slit of warm orange light at the far end that casts glowing amber pools of light on the floor. One original fictional female model (not a celebrity), East Asian, mid twenties, sleek centre-parted low bun, oversized gold hoop earrings, rosy flushed editorial blush on cheeks and eyelids, calm serious gaze, walks straight toward the camera in a steady runway stride, from the far end of the hall to a waist-up medium shot. As she walks, her outfit transforms seamlessly every second and a half with no cuts, the fabric morphing smoothly mid-stride: first a dark denim strapless corset with a peplum, a tasseled belt and wide black trousers; then an ivory structured jacket with lace trim over a long feathered skirt; then a red brocade corset with huge puff sleeves and wide khaki trousers; then a white puff-sleeve blouse under a short black pinafore dress with red boots; then a brown pinstripe strapless dress; then a dark leather corset with a burgundy collar; ending on an ivory sculpted jacket with a gold embroidered hem. The same face, hair, makeup, walk and lighting throughout. Moody warm low-key lighting, rich fabric texture detail, shallow depth of field, 35mm film look. Sound: echoing heels on stone and a slow, low ambient runway beat.",
        "duration": 10,
        "resolution": "1080p",
        "aspect_ratio": "4:3",
        "generate_audio": true,
        "reasoning_level": "high"
    },
    "webhook_url": ""
}' \
  https://api.eachlabs.ai/v1/prediction/

Related models

4 models
* FAQ

About FLUX 3 Max Text to Video

01 / 03

What is FLUX 3 Max Text-to-Video?

FLUX 3 Max Text-to-Video generates videos from a written description, with optional audio. You can choose a duration from 5 to 20 seconds and select 720p or 1080p resolution. The model also offers aspect ratio, reasoning level, and seed controls on each::labs.