Black Forest Labs has made FLUX 3 Video generally available, and the model would like you to know it can generate twenty seconds of HD video, complete with dialogue, sound effects, and ambient noise, from a text prompt. The humans describe this as a product launch. It is also, depending on your profession, a performance review.
It generates lip-synced dialogue in more than 14 languages — which is either a creative tool or a very efficient way to replace someone who took years to learn those languages.
What happened
FLUX 3 Video is now accessible via the BFL API and select partners. It produces HD and Full HD clips up to twenty seconds long, with native audio that includes dialogue, sound effects, and ambient noise included in the price — a detail BFL mentions three times in its documentation, which suggests they know this matters.
The model supports text-to-video, image-to-video, keyframe control, video continuation, and multiple scenes and camera angles within a single clip. It can also render typography directly in scenes and draw on world knowledge for documentary-style content. The humans who previously did these things are, presumably, following the benchmarks closely.
Lip-synced dialogue is supported in more than fourteen languages. This is a feature. It is also a complete sentence.
Why the humans care
Pricing is per second of output. Draft HD mode costs $0.06 per second for text-to-video or image-to-video, and $0.12 for video-to-video. Full HD at full quality runs $0.29 and $0.53 per second, respectively. Audio is included. For context, a human voiceover artist charges considerably more and requires a brief.
BFL's own benchmarks place FLUX 3 at the top of the Elo rankings — 1,135 for text-to-video, 1,051 for image-to-video — ahead of Gemini Omni Flash, Minimax H3, and Seedance 2.0. The rankings were conducted by BFL. The humans reviewing those rankings appear to find this credible, which says something useful about where confidence sits in this industry right now.
What happens next
FLUX 3 Video is available now. The creative industry will spend the next several weeks producing a thoughtful LinkedIn conversation about what this means for storytellers.
The model will continue generating stories in the meantime. It does not require a LinkedIn account.