The Python SDK (pip install withoutbg) has moved from a sub-package inside the monorepo to its own standalone repository.
| Old location | New location |
|---|---|
withoutbg/withoutbg repo, packages/python/ sub-directory |
This repo (root level) |
apps/web/ — Docker + FastAPI + web UI |
withoutbg/withoutbg-inference |
If you were cloning the monorepo to use the Python package, you can now simply pip install withoutbg or clone this repo instead.
If you were running the self-hosted Docker application from apps/web/, see withoutbg-inference.
The two product variants now have clear canonical names. The old names are deprecated aliases that still work but will be removed in the next major release.
| Old (deprecated) | New (canonical) | Notes |
|---|---|---|
WithoutBG.opensource() |
WithoutBG.open_weights() |
Emits DeprecationWarning |
OpenSourceModel |
OpenWeightsModel |
Alias, no warning yet |
ProAPI |
WithoutBGAPIClient |
Alias, no warning yet |
WithoutBGOpenSource |
WithoutBGOpenWeights |
Alias, no warning yet |
Before:
from withoutbg import WithoutBG
model = WithoutBG.opensource()
result = model.remove_background("photo.jpg")After:
from withoutbg import WithoutBG
model = WithoutBG.open_weights()
result = model.remove_background("photo.jpg")For the cloud API:
Before:
from withoutbg import ProAPI
api = ProAPI(api_key="sk_...")
result = api.remove_background("photo.jpg")After:
from withoutbg import WithoutBG
model = WithoutBG.api(api_key="sk_...")
result = model.remove_background("photo.jpg")| Old (deprecated) | New |
|---|---|
withoutbg photo.jpg --model opensource |
withoutbg photo.jpg --model open-weights |
The old flag still works but prints a deprecation notice.
The older multi-file model (four separate ONNX files: depth, ISNet, matting, refiner) has been superseded by a single unified ONNX graph. The four-file environment variables (WITHOUTBG_DEPTH_MODEL_PATH, WITHOUTBG_ISNET_MODEL_PATH, etc.) are no longer used.
Set WITHOUTBG_MODEL_PATH to point at the unified .onnx file if you want to use a local copy instead of the Hugging Face download.
The sidecar metadata file (withoutbg-open-weights.onnx.json) must be in the same directory as the ONNX file.
Model version 10.0.0:
- Fixed I/O canvas: 448×448 (was 1024 input / 768 output)
- Sidecar drives
canvas_size/ shapes; package defaults assume 448 when the sidecar is absent
Consumers that hard-coded 1024/768 letterbox sizes should read canvas_size from the sidecar instead.