Flux 3 AI video generator

Black Forest Labs' Flux 3. 20-second video with native audio.

Every video model

One workspace

Made with Flux 3

Key features

Generate a video in three steps

  1. 01

    Open Morphic

    Sign up and start generating on a free-flowing, infinite visual canvas.

  2. 02

    Describe the shot

    Subject, action, camera move, light, and one line about what it should sound like.

  3. 03

    Generate and refine

    Draft a direction first, then render it at full quality and compare takes side by side.

What Flux 3 generates

Video and sound in one pass

Dialogue, effects, ambience, and music are generated with the frames, not added after. A hammer strike rings on the hit and a spoken line already matches the mouth, so a clip is usable the moment it lands.

Twenty seconds, seven aspect ratios

Render 5 to 20 seconds at HD or Full HD, framed anywhere from 21:9 cinemascope to 9:16 vertical. One prompt covers the cinema cut and the phone cut without a reframe pass.

Keyframes you pin yourself

Give Flux 3 one still for the opening frame, two to fix a start and an end, or up to ten pinned to timestamps. The model interpolates one continuous shot that has to hit every mark in order.

Continue a clip you already have

Feed in existing footage and Flux 3 picks up from its final frames, carrying momentum, camera behavior, and the audio bed across the seam. The scene keeps going instead of starting over.

Dialogue in a dozen languages

Quote the line, name the language, and describe the delivery. Flux 3 speaks it on camera with matching accent and lip-sync across English, Chinese, Spanish, French, German, Japanese, Hindi, and more.

Type that survives the motion

Titles, signage, and lower-thirds render as part of the scene and stay legible while the camera moves, so a kinetic title card or a product label holds up without a separate motion-graphics pass.

All on Morphic

Your complete video stack

Simple pricing

Get started for free today, with the option to upgrade or cancel anytime.

Basic

$9/ month
billed as $0 per year

900 monthly credits

1 user only

All models

Workflows

Standard

$24/ month
billed as $0 per year

3200 monthly credits

1 user only

All models

Workflows

Pro

$45/ month
billed as $0 per year

6200 shared monthly credits

1 user

+ up to 4 more at extra cost

All models

Workflows

Pro Max

$170/ month
billed as $0 per year

24000 shared monthly credits

1 user

+ up to 9 more at extra cost

All models

Workflows

Enterprise

For higher limits

Custom

pricing and billing terms

High-volume credits
Custom seat limits
All models
Workflows
Pricing Gradient

Free

For playing around

$0

forever free

Up to 20 credits
1 user only
Limited models
Workflows

FAQs

What is the Flux 3 AI video generator?
Flux 3 is Black Forest Labs' first video model, built on a multimodal foundation that learns image, video, and audio together. You describe a shot and it renders the moving frames and the matching soundtrack in the same pass, up to 20 seconds long. Black Forest Labs made video generation generally available on August 4, 2026.
How long can a Flux 3 video be?
Flux 3 generates 5 to 20 seconds in a single pass, and it can also pick the length automatically to suit the prompt. Twenty seconds covers a full ad spot or a short scene in one generation, rather than cutting several short clips together afterward.
What resolution does Flux 3 output?
Flux 3 renders at HD 720p by default and Full HD 1080p through upscaling. It supports seven aspect ratios, from 21:9 cinemascope through 16:9, 2:1, 4:3, 1:1, and 3:4 down to 9:16 vertical, so the same prompt can be framed for a cinema screen or a phone.
Does Flux 3 generate audio with the video?
Yes, and it is on by default. Dialogue, sound effects, ambience, and music are generated alongside the frames, so an impact or a spoken line is already in sync. Name the sound you want in the prompt. You can also switch audio off for a silent clip.
What modes does the Flux 3 video generator run?
Three. Text-to-video renders a shot from a written prompt alone. Image-to-video animates stills you supply, either as the opening frame or as pinned keyframes. Video continuation takes an existing clip and carries its motion, framing, and audio forward without a cut.
How do Flux 3 keyframes work?
You pin stills and Flux 3 interpolates one continuous shot through them. A single image becomes the opening frame, a pair fixes the start and the end, and up to ten images pinned to timestamps turn the clip into a storyboard the model has to hit in order.
Can Flux 3 extend a video I already have?
Yes. Video continuation accepts an existing clip and picks up from its final frames, carrying momentum, camera behavior, dialogue, and audio across the seam. It reads both the picture and the sound of the source, so the continuation matches rather than restarting the scene.
What is Flux 3 draft mode?
Draft mode returns a fast, low-cost preview of your prompt so you can explore directions instead of waiting on a full render each time. When a draft looks right, Flux 3 renders it at full quality with the same subjects, composition, and motion, so the final matches the version you approved.
Can Flux 3 generate dialogue in other languages?
Yes, with lip-sync. Supported languages include English and its dialects, Chinese, Spanish, French, German, Japanese, Portuguese, Russian, Italian, Indonesian, Turkish, Hindi, and Punjabi. Quote the exact line in the prompt, name the language, and describe the delivery.
Can Flux 3 put several shots in one video?
Yes. Write the prompt as ordered shots with hard cuts between them and Flux 3 blocks multiple scenes and camera angles inside a single generation, holding character and look across the cuts while one audio bed carries through the whole clip.
Can I use the Flux 3 AI video generator online?
Yes. Everything runs in the browser on Morphic, with no download or install. Describe the shot, generate, and the clip lands on your Canvas from any laptop, including machines with no GPU.
Can I use Flux 3 on Morphic?
Yes. Flux 3 runs on Morphic today, next to video models like Veo, Kling, Seedance, and Vidu, and to the rest of the Flux family for images. That means you can generate the same shot on two models and keep the take that lands, all on one Canvas.

You might also like