Inpainting on FLUX.1 Fill [dev], Black Forest Labs' 12-billion-parameter fill model, with up to three of your own FLUX.1 dev LoRAs. Send a source image and a mask: only the white area is redrawn, the rest comes back as it was, and nothing is filtered.
Workbench
These examples were written for this endpoint. Replace the prompt and sample media URLs with your own accessible inputs, then check the current quote and model availability before submitting.
Outputcrimson-slip-fitting The dress and its reflection in the left mirror panel were masked; the face, arms and room are the original pixels. 1.5k tier.
- 42
- 1.5k
Outputhoodie-unzipped-studio Only the front of the hoodie and its reflection were masked, so the hood carries over and the edit reads as the same garment.
- 7
Outputpixel-mural-lora Two wall rectangles masked, with a pixel-art LoRA at 1 and its trigger word first. Compare with the run below on the same seed.
- 5
Outputpixel-mural-no-lora The same image, mask, prompt and seed without the LoRA: the mural is flatter and the dancers simpler.
- 5
This example starts at $0.042 per image. The Playground updates the estimate when you change the settings.
