FLUX learns object erasure as a single model-level task
- Source
- x.com
- Date
New Release: Erase Remove objects, erase text, and clean up details in any image. FLUX is now trained to erase and reconstruct as one task at the model level. We noticed this approach leads to images that stay clean and consistent. Give it an image + a simple “erase mask” over what you want gone, and it reconstructs the background - shadows and missed bits included - while keeping the rest intact. Try the demo: https://t.co/1iWnhtFb1F Available on the BFL API:
Context
FLUX Erase turns object removal into a task defined by an image and a marked region. The user supplies a mask identifying what should disappear; the model generates replacement background content. Black Forest Labs says it trained erasure and reconstruction together, including removal of associated details such as shadows. That targets the cleanup left behind when an object disappears but its visual traces remain. The API documentation specifies a black-and-white mask and says no written prompt is required, while the launch thread explains the training approach.
Checking sign-in…
Loading comments…