FLUX 3 Image is now available, supporting layout framing, generation, and 10 reference images
BIBIBI
AT A GLANCE
According to the report,FLUX 3 Image supports layout framing, image generation and editing, and output up to.
Article
FLUX 3 Image is now available, enabling users to frame the composition first and then generate the image, with support for up to native and 10 reference images
Dongcha Beating AI news flash,Black Forest Labs has released FLUX 3 Image, supporting image generation and editing. Users can first use a Bounding Box to define the position and size of elements, then have the model generate an image according to this layout.
For example, when creating a poster with multiple people, users can first frame each person's position and then specify who stands where. The model can also plan these boxes itself. After generation, users can continue moving or modifying elements while minimizing the impact on other parts of the image.FLUX 3 Image supports up to 10 reference images and can generate native and images. Enterprises can already purchase the model weights for self-hosted deployment and fine-tuning. An open-weights version for general users will be released in the coming weeks.
Original link https://m.theblockbeats.info/flash/370079
Key points
01
According to the report,Black Forest Labs has released FLUX 3 Image。
02
FLUX 3 Image supports image generation and editing.
03
Users can use bounding boxes to define the position and size of elements, then generate an image according to the layout.
04
The model can plan bounding boxes itself.
05
After generation, users can continue moving or modifying elements while minimizing the impact on other parts of the image.
06
Supports up to 10 reference images.
07
It can generate native and images.
08
Enterprises can currently purchase the model weights for self-hosted deployment and fine-tuning.
09
An open-weights version for general users will be released in the coming weeks.
AI-assisted interpretation
The following is analysis, separate from reported facts. Verify important claims independently.
This is a model for generating and editing images. Users can first arrange the position and size of elements in the image using bounding boxes, and then generate the image; after generation, they can continue adjusting the elements. The report also mentions support for multiple reference images, native and output, as well as enterprise self-deployment and fine-tuning.
Why it matters to readers
If the report is accurate, users will be able to control image layouts more precisely, and enterprises will also be able to purchase the model weights for deployment and fine-tuning.
A bounding box can be understood as first outlining on the canvas where a person or object should appear, and then having the model complete the image.
Risks and unknowns
Not included in the evidence Black Forest Labs official announcement or technical specifications.
The evidence does not specify the conditions for purchasing, deploying, and fine-tuning enterprise model weights.
The evidence does not specify the actual image quality, functional availability, or editing results.
Whether the open-weight version will be released in the coming weeks remains to be confirmed.
Related Developments
Loading event timeline…
Related concepts
Bounding Box
This term is not in the glossary yet. Browse related concepts in the glossary.
