Registry / ai-ml / onnxsim-prebuilt

onnxsim-prebuilt

JSON →
library0.4.39.post2pypypiunverified

Prebuilt wheels of onnxsim (ONNX Simplifier) distributed as a standalone pip package for platforms that lack a compatible compiler. Current version is 0.4.39.post2, with irregular releases tied to upstream onnxsim updates.

pip install onnxsim-prebuilt
INSTALL
IMPORT
SIG · ONNXSIM-PREBUILT
O
onnxsim-prebuilt
ai-mlpythonv0.4.39.post2
Install
7.8s avg
Import
3443ms
Disk
289MB
Pass rate
4/ 10
Env Coverage4 / 10
glibc
3.93.13
musl
3.93.13
Install & Compatibility
Where this runs
tested against v0.4.39.post2 · pip install
no network on importno background threads
Install × environment matrix
Each cell = how many times install + import succeeded across repeated harness runs. Partial = flaky.
glibc = Debian/Ubuntu slim · musl = Alpine Linux
musl
glibc
py 3.10
✕ build_error
✓ 8s
py 3.11
✕ build_error
✓ 7.9s
py 3.12
✕ build_error
✓ 7.5s
py 3.13
✕ build_error
✓ 7.6s
py 3.9
✕ build_error
✕ build_error
289MB installed
● package 289MB
Code
Verified usage

Verified import paths — ran on the pinned version, not inferred.

simplify
from onnxsim import simplify
from onnxsim_prebuilt import simplify
The package name on pypi is onnxsim-prebuilt, but the import module is still onnxsim.

Basic usage: load a model, simplify, save.

import onnx from onnxsim import simplify # Load your ONNX model model = onnx.load('model.onnx') # Simplify the model model_simp, check = simplify(model) if check: onnx.save(model_simp, 'model_simplified.onnx') print('Simplified model saved.') else: print('Simplification failed to preserve outputs.')
onnxsim --version
Debug
Known issues
gotchaDo NOT import from 'onnxsim_prebuilt'. The import module is 'onnxsim', matching the upstream package. The 'onnxsim-prebuilt' name only applies to the pip package.
fix
Use 'from onnxsim import simplify' instead of 'from onnxsim_prebuilt import simplify'.
affects: all
gotchaThe 'simplify' function may produce a model that does not preserve output exactness for all ops. Always check the returned boolean flag.
fix
After calling simplify, verify that the returned 'check' is True before using the simplified model.
affects: all
deprecatedThe function 'check_out' (if used) is deprecated in favor of the built-in check in simplify.
fix
Use the return value of simplify instead of calling check_out separately.
affects: >=0.4.36
Upgrade
Version history
0.4.39.post2latest on PyPI · released Jan 29, 2026
Audit
Dependencies
onnxrequiredRequired to load and save ONNX models
onnxsimrequiredCore library; onnxsim-prebuilt is a packaging of onnxsim with prebuilt wheels
Agent activity
2 hits · last 30 days
node
2
Resources
onnxsim-prebuilt — pip install onnxsim-prebuilt · libregistry