Video teams wanting clips with native sound and lip sync via API now have it generally available, at up to $0.80 a second for UHD.

Kenning's take

What it does

FLUX 3 is a multimodal model from Black Forest Labs that learns from images, video and audio at once. Its first public use is FLUX 3 Video, which makes clips of up to 20 seconds with native audio and lip-synced dialogue in many languages, from text, images, keyframes or an existing clip. It was announced in early access in July 2026, and video generation became generally available on the BFL API on 4 August 2026.

• Text-to-video and image-to-video with audio
• Clips up to 20 seconds at 720p or 1080p
• Multilingual dialogue with lip-sync
• Video continuation and keyframes
• Multi-shot scenes in one generation
• Cheap draft mode

Who it suits. Developers and video teams generating video with sound through an API.

Key facts

Category

AI models

Made by

Black Forest Labs

Released

23 July 2026

How to get it

BFL API and Playground, plus select partners; video generally available since 4 August 2026

Works on

API, web playground

API price

Text or image to video $0.17/s (hd), $0.29/s (fhd), $0.40/s (qhd), $0.80/s (uhd); draft $0.06/s; continuation from $0.41/s

Models and features

• FLUX 3 Image 2026-10-01

FAQ

Facts from the maker's own pages and the sources listed. Checked 2026-10-11.