Black Forest LabsVideo model

FLUX 3 Video

Picture and sound, together.

FLUX 3 Video sample output
Full HD5 to 20sSynced audio
From 105 credits per 5 secondsGenerate with FLUX 3 Video

Why FLUX 3 Video

Sound that was never separate

Picture and audio come out of one model, so the footstep lands on the footstep. Dialogue, effects and room tone, generated together and included in the price.

Made for

  • Paid social ads

    Full HD with a voice, a hook and sound design, ready to upload. No edit suite between the prompt and the ad.

  • Spokesperson and UGC

    A presenter who speaks your script, lip synced, in the language your market actually buys in.

  • Product launches

    Twenty seconds at Full HD with typography that animates, which is enough for a launch film.

  • Localisation at volume

    Re-render the same spot in eight languages from eight prompts instead of eight recording sessions.

  • Concepting

    Draft mode is cheap enough to be wrong in public. Run the whole idea list, keep one.

The specs

Provider
Black Forest Labs
Resolution
HD up to 1MP per frame, Full HD up to 2MP per frame
Frame rate
24fps
Duration
5 to 20 seconds
Audio
Native and synchronized, included at no extra cost
Dialogue
Multilingual, with lip sync
Modes
Text to video, image to video, video to video, keyframes
Keyframes
Up to 10 per generation, including start and end frames
Aspect ratios
Auto, 21:9, 2:1, 16:9, 4:3, 1:1, 3:4, 9:16
Cost
From 105 credits per 5 seconds, or 31 in Draft

Straight answers

What is FLUX 3 Video?

FLUX 3 Video is Black Forest Labs' video generation model, launched in July 2026. It produces 5 to 20 second clips at HD or Full HD with a native synchronized audio track, from text, an image, keyframes or an existing video.

Does the audio cost extra?

No. Audio is generated with the picture and included in the credit cost. There is no separate voice or sound effects charge.

What is Draft mode?

FLUX 3 Video Draft is the same model rendering at HD for roughly a third of the price, from 31 credits per 5 seconds. It exists so exploration is cheap. Once a take works you re-render it in Full HD.

Can it speak languages other than English?

Yes, with lip sync. Chinese, Spanish, French, German, Japanese, Portuguese, Russian, Italian, Indonesian, Turkish, Hindi and Punjabi are all supported, among others.

How much does FLUX 3 Video cost on APImage?

From 105 credits per 5 seconds at HD, rising with length, Full HD and video to video. Draft starts at 31. The exact cost of your settings is shown before you generate.

How does it compare to Seedance?

FLUX 3 Video is the one for anything going public: it is the only model here that renders sound with the picture and it reaches Full HD. Seedance 2.0 is cheaper for silent 1080p motion work, and Seedance 2.5 goes to 30 seconds where FLUX 3 stops at 20.

How do I use it?

Open the video studio and pick FLUX 3 Video, or FLUX 3 Video Draft to explore cheaply. Write the shot and the sound. Say the line you want spoken and the ambience you want under it. Add keyframes if the timing matters, up to ten including the start and end frames. Pick HD or Full HD, generate, and upgrade the take you keep.

FLUX 3 Video
is live.

Generate with FLUX 3 Video
From 105 credits per 5 secondsOne credit poolNo per-model surchargeAPI and MCP