Black Forest Labs 正式发布 FLUX 3 Video，称其超越 Seedance 2.0

AI公司黑森林实验室（Black Forest Labs，简称BFL）已通过BFL API和部分合作伙伴，让其FLUX 3视频模型普遍可用。该模型可以生成最长20秒的高清（HD）和全高清（Full HD）视频短片，带有原生音频，包括对话、音效和环境噪音。它支持文本到视频、图片到视频、关键帧、视频续拍，以及单个短片内的多场景和多角度镜头。BFL表示，FLUX 3可以在场景中直接渲染排版，遵循复杂提示，并利用世界知识进行例如纪录片的创作。它还可以生成超过14种语言的唇形同步对话。

BFL自测显示，FLUX 3在文本到视频的Elo排名中得分1135，在图片到视频中得分1051，位居榜首。它领先于Gemini Omni Flash：https://the-decoder.com/google-launches-nano-banana-2-lite-for-fast-ai-images-and-gemini-omni-flash-for-video-via-api/，Minimax H3：https://the-decoder.com/chinas-minimax-h3-is-the-first-open-model-to-top-an-ai-video-ranking/，以及Seedance 2.0：https://the-decoder.com/hollywood-wants-seedance-banned-and-reportedly-also-wants-to-keep-using-it/。

定价：https://bfl.ai/pricing 基于每秒视频输出计费。草稿模式仅限HD，文本到视频或图片到视频费用为每秒0.06美元，视频到视频为每秒0.12美元。全质量模式下，HD文本到视频或图片到视频为每秒0.17美元，视频到视频为每秒0.41美元。全高清（Full HD）分别为每秒0.29美元和0.53美元。音频包含在内。更多视频示例可见BFL博客：https://bfl.ai/blog/flux-3-video。广告 DEC_D_Incontent-1 广告

保持对AI的关注。清晰、有用，无废话。

关注The Decoder，获取AI新闻、背景故事和专家分析。

解码器：https://the-decoder.com/

AI company Black Forest Labs (BFL) has made its FLUX 3 Video model generally available through the BFL API and select partners. The model generates HD and Full HD clips up to 20 seconds long, with native audio that includes dialogue, sound effects, and ambient noise. It supports text-to-video, image-to-video, keyframes, video continuation, and multiple scenes and camera angles within a single clip. BFL says FLUX 3 can render typography directly in scenes, follow complex prompts, and draw on world knowledge for uses such as documentaries. It also generates lip-synced dialogue in more than 14 languages.

BFL's own tests put FLUX 3 at the top of the Elo rankings for text-to-video with a score of 1,135 and image-to-video with 1,051. It ranks ahead of Gemini Omni Flash：https://the-decoder.com/google-launches-nano-banana-2-lite-for-fast-ai-images-and-gemini-omni-flash-for-video-via-api/, Minimax H3：https://the-decoder.com/chinas-minimax-h3-is-the-first-open-model-to-top-an-ai-video-ranking/, and Seedance 2.0：https://the-decoder.com/hollywood-wants-seedance-banned-and-reportedly-also-wants-to-keep-using-it/.

Pricing：https://bfl.ai/pricing is based on each second of video output. Draft mode is limited to HD and costs $0.06 per second for text-to-video or image-to-video, and $0.12 for video-to-video. At full quality, HD costs $0.17 per second for text-to-video or image-to-video, and $0.41 for video-to-video. Full HD costs $0.29 and $0.53 per second, respectively. Audio is included. More video examples are available on the BFL blog：https://bfl.ai/blog/flux-3-video. Ad DEC_D_Incontent-1 Ad

Stay in the loop on AI. Clear, useful, no fluff.

Follow The Decoder for AI news, background stories and expert analyses.

The Decoder：https://the-decoder.com/