Skip to main content
πŸ“– The AI Tool Bible
ONNX preview image
ONNX logo

ONNX

βœ“ Text source-checked Nov 22, 2026

ONNX is an open format built to represent machine learning models.

Source-checked against the vendor's site Β· rebuilt from the vendor's own pagesProfile updated Link checked How we rate

FreeFine-tuning7.0 / 10

ONNX is an open format built to represent machine learning models. ONNX defines a common set of operators and a common file format to enable AI developers to use models with a variety of frameworks, tools, runtimes, and compilers. ONNX provides an open source format for AI models, both deep learning and traditional ML.

ONNX enables developers to use their preferred framework with their chosen inference engine. ONNX makes it easier to access hardware optimizations.

ONNX is a community project.

ONNX released packages are published in PyPi. The ONNX package provides abi3-compatible wheels, allowing a single binary wheel to work across multiple Python versions from 3.12 onwards.

From the vendor's pages

According to ONNX: ONNX is an open format built to represent machine learning models. Its pages list this strength: ONNX enables the use of preferred frameworks with chosen inference engines.

β€” The AI Tool Bible editorial team

Pros

  • βœ… ONNX enables the use of preferred frameworks with chosen inference engines.
  • βœ… ONNX makes it easier to access hardware optimizations.
  • βœ… ONNX is a community project.

Facts

Platforms
linux, api (not source-checked)
API available
Yes (not source-checked)
Open source
Yes (license: Apache-2.0)
GitHub stars
21,675 (checked Nov 13, 2026) (not source-checked)
Last GitHub push
Nov 13, 2026 (not source-checked)
First commit
Sep 2017 (not source-checked)
Package downloads (30 d)
17,129,863 Β· pypi onnx Β· checked Nov 19, 2026 (not source-checked)
Hacker News
9 stories Β· top 3 points Β· last Sep 4, 2022 Β· checked Nov 19, 2026
Hugging Face
Author onnx-community Β· checked Nov 17, 2026

Where this tool sits

Use cases

representing machine learning modelsframework interoperabilityhardware optimization access
See alternatives to ONNX β†’

Explore related

Reviews