You're looking at a specific version of this model. Jump to the model overview.

zf-kbot /object-remover-20251030:77061e4f

Input schema

The fields you can use to run this model with an API. If you don’t give a value for a field its default value will be used.

Field Type Default value Description
image
string
Base image to edit.
mask
string
Mask image (white = area to edit).
preprocessing_resolution
None
720p
Downscale images to target resolution for faster inference (original = no downscaling)
num_inference_steps
integer
10

Min: 1

Max: 100

Number of diffusion steps.
prompt
string
Remove the object and restore background seamlessly.
Positive prompt.
negative_prompt
string
text, watermark, logo, artifacts, patterns, distorted edges, blurry texture, extra object, unnatural shadow, duplicated area, unrealistic lighting
Negative prompt used for guidance.
task
None
object_removal
Task preset (auto fill prompt if blank).
transformer_variant
None
one_reward
Choose the released OneReward transformer.
guidance_scale
number
1.4

Max: 20

Classifier-free guidance scale.
true_cfg
number
4

Max: 10

True-CFG weight inside pipeline.
seed
integer
0
Random seed (-1 for random).
apply_object_removal_lora
None
1
Enable object-removal LoRA ("0" = off, "1" = on).
lora_weight
number
1

Max: 2

LoRA adapter weight.
enable_crop_optimization
None
disabled
Crop to masked region for faster inference
crop_padding
integer
128

Max: 512

Padding around masked region when cropping (pixels)
auto_config_mode
None
disabled
Auto-tune object removal parameters based on mask and presets.
auto_remove_type
None
general
Content type being removed when auto config is enabled.
auto_quality_mode
None
balanced
Quality profile for auto configuration.
auto_mask_ratio_override
number
-1

Min: -1

Max: 1

Override mask area ratio (0-1). Set <0 to use auto-computed ratio.
auto_config_overrides
string
JSON overrides for auto configuration (advanced).
log_level
None
debug
Log level: 'info' for key info, 'debug' for detailed logs

Output schema

The shape of the response you’ll get when you run this model with an API.

Schema
{'format': 'uri', 'title': 'Output', 'type': 'string'}