unifyai/ivy

Health v2 · maintenance only

Flags: no_license

Adoption not part of the score

14172 stars · 5457 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

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

library · maturity active

machine-learning compiler interpreter developer-tools machine-learning deep-learning developer-tools python cross-platform transpiler framework-agnostic pytorch tensorflow jax numpy code-conversion algorithms

Member repositories

For agents

markdown · JSON · MCP: product_card(name="unifyai/ivy")