started · updated
Black Forest Labs releases new FLUX 3 AI video and image models
Black Forest Labs and fal.ai have launched FLUX 3 Max, a video generation model capable of producing 10-second clips in approximately 12.45 seconds. The model utilizes post-training techniques to improve prompt adherence and visual quality, building upon the original FLUX 3 base. It offers various API input modes, including text, image, and keyframe-based control.
In the realm of image generation, Black Forest Labs released FLUX 3 Image Klein as an open-weight model. This 5.6B parameter model supports text-to-image, image editing, and multi-reference generation. Notably, the INT8 version is optimized to run on consumer hardware with as little as 8GB of VRAM.
Additionally, SYNZIFY AI TECH LLC has launched kling4.kr, an independent AI video platform tailored for the Korean market. The service features 4K video rendering up to 30 seconds and includes tools for multi-reference control and lip-syncing to assist creators and marketers.
Entities
Claims
What the coverage asserts, and how many sources carry each claim.
- [○ 1 SOURCE] The FLUX 3 Image Klein model can run on hardware with 8GB of VRAM using the INT8 version. pc.watch.impress.co.jp
- [○ 1 SOURCE] SYNZIFY AI TECH LLC is operating the kling4.kr platform for Korean users. meeco.kr
- [○ 1 SOURCE] FLUX 3 Max can generate a 10-second video in an average of 12.45 seconds. www.tokenpost.kr
- [○ 1 SOURCE] Black Forest Labs released the FLUX 3 Image Klein open-weight image generation model. pc.watch.impress.co.jp
- [○ 1 SOURCE] fal.ai and Black Forest Labs released the FLUX 3 Max video generation model. www.tokenpost.kr