Home
Black Forest Labs releases Flux3, first multimodal model with native 20-second audio-visual sync
On July 23, German AI startup Black Forest Labs officially launched the multimodal foundation model Flux3, marking a significant leap forward in generative video technology. Built on the Self-Flow architecture, the model integrates dedicated codecs for image, video, audio, and motion, enabling unified understanding and generation across both physical and digital environments.

As the first model to support native audio generation, Flux3 can produce synchronized audio-video clips of up to 20 seconds, covering capabilities such as text/image/video-to-video conversion, keyframe-based transitions, and multilingual dialogue. In early benchmarks at 720p resolution for 10-second clips, Flux3 showed strong competitiveness, outperforming Luma Ray3.2 (93% win rate) and Runway Gen-4.5 (77% win rate), while maintaining a slight edge over top-tier models like Seedance2.0 and Gemini Omni Flash.
In addition, the company developed the video action model Flux-mimic in collaboration with Mimic Robotics for the robotics sector, which has already begun production task testing at an Audi factory. BFL has adopted a phased release strategy: Flux3Video is available now, with Flux3Image and the open-source weight "Flux3Dev" to follow shortly. By overcoming the limitations of single-modal information, Flux3 is accelerating AI's progress toward a "world model" capable of perception and action in the physical world.
Related article
Calibre Hands-On AI Interactive Writing: The E-book Manager Turns into a Text Adventure Game Engine
Calibre, the open-source e-book management tool, has released version 9.15, introducing a creative new feature: an optional AI-powered interactive writing game called "Create Your Own Adventure."By default, this game is hidden and does not appear whe
Apple Unveils First Public Beta of iOS 27, Marking Siri’s Major AI Overhaul
Apple has officially released the first public beta version of iOS 27 for users today. This update does not require a developer account. Users simply need to register or log in on the Apple Beta Software Program website and select the corresponding v
OpenAI Adds GPT-Live-1 to API With Full-Duplex Voice and Interruption Support for Customer Service Calls
OpenAI has integrated GPT-Live-1 into its API, empowering developers to create applications and business workflows that enable real-time voice interactions. This model facilitates full-duplex conversations, allowing it to listen and speak simultaneou
Related Special Topic Recommendations
Comments (0)
0/500
On July 23, German AI startup Black Forest Labs officially launched the multimodal foundation model Flux3, marking a significant leap forward in generative video technology. Built on the Self-Flow architecture, the model integrates dedicated codecs for image, video, audio, and motion, enabling unified understanding and generation across both physical and digital environments.

As the first model to support native audio generation, Flux3 can produce synchronized audio-video clips of up to 20 seconds, covering capabilities such as text/image/video-to-video conversion, keyframe-based transitions, and multilingual dialogue. In early benchmarks at 720p resolution for 10-second clips, Flux3 showed strong competitiveness, outperforming Luma Ray3.2 (93% win rate) and Runway Gen-4.5 (77% win rate), while maintaining a slight edge over top-tier models like Seedance2.0 and Gemini Omni Flash.
In addition, the company developed the video action model Flux-mimic in collaboration with Mimic Robotics for the robotics sector, which has already begun production task testing at an Audi factory. BFL has adopted a phased release strategy: Flux3Video is available now, with Flux3Image and the open-source weight "Flux3Dev" to follow shortly. By overcoming the limitations of single-modal information, Flux3 is accelerating AI's progress toward a "world model" capable of perception and action in the physical world.
Calibre Hands-On AI Interactive Writing: The E-book Manager Turns into a Text Adventure Game Engine
Calibre, the open-source e-book management tool, has released version 9.15, introducing a creative new feature: an optional AI-powered interactive writing game called "Create Your Own Adventure."By default, this game is hidden and does not appear whe
Apple Unveils First Public Beta of iOS 27, Marking Siri’s Major AI Overhaul
Apple has officially released the first public beta version of iOS 27 for users today. This update does not require a developer account. Users simply need to register or log in on the Apple Beta Software Program website and select the corresponding v
OpenAI Adds GPT-Live-1 to API With Full-Duplex Voice and Interruption Support for Customer Service Calls
OpenAI has integrated GPT-Live-1 into its API, empowering developers to create applications and business workflows that enable real-time voice interactions. This model facilitates full-duplex conversations, allowing it to listen and speak simultaneou











