# unifyai/ivy

Convert Machine Learning Code Between Frameworks

Repository: https://github.com/unifyai/ivy
Canonical: https://ross.abutalabs.com/products/ivy
Homepage: https://ivy-llc.github.io/docs/
Language: Python
License: NOASSERTION
License Family: other
Topics: python, tensorflow, pytorch, numpy, jax
Last push: 2026-08-11T19:06:27+00:00

## Health v2 (maintenance only)
Score: 78/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 97, release rhythm 40, longevity 100
- inputs: {"age_days": 2052, "days_push": 22, "days_rel": 444, "gap_med": 29, "n_releases_24m": 6}
- flags: no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 14172, forks 5457 (observed 2026-08-28T04:11:05.743470+00:00)

## What it is
Ivy is a Python library that transpiles machine learning code, models, and libraries between frameworks such as PyTorch, TensorFlow, JAX, and NumPy. It provides a unified API so code written for one framework can run on another via ivy.transpile.

## Use cases
- convert a pytorch model to tensorflow
- transpile ML code between frameworks
- run jax code with pytorch backend
- port a library from tensorflow to numpy
- make my ML code framework-agnostic
- convert deep learning models between frameworks

## When to choose
- you need to use a model or library written for a different ML framework than your stack
- you want to avoid rewriting framework-specific code manually
- you need a unified API across PyTorch, TensorFlow, JAX, and NumPy

## When to avoid
- your code already targets a single framework and you never need to switch
- you need maximum performance with zero transpilation overhead
- you rely on very new or exotic framework features that transpilation may not support

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, compiler, interpreter, developer-tools
- domain: machine-learning, deep-learning, developer-tools
- platform: python, cross-platform
- tags: transpiler, framework-agnostic, pytorch, tensorflow, jax, numpy, code-conversion, algorithms

## Member repositories
- unifyai/ivy (main) score 78

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:11:05.743470+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T17:12:45.243727+00:00, confidence not recorded.
  - readme: https://github.com/unifyai/ivy (fetched 2026-08-28T04:11:05.743470+00:00, sha 41a3d67a25b8)
- Data as of 2026-08-30T08:39:29.467469+00:00.