Autodiff Python Packages

uncertainties

Transparent calculations with uncertainties on the quantities involved (aka "error propagation"); calculation of derivatives.

2.2M 680 88

autograd-gamma

NotImplementedError: VJP of gammainc wrt argnum 0 not defined

1.7M 15 4

advect

Automatic differentiation for the NumPy code you already write

11K 9 1

pennylane-catalyst

A JIT compiler for hybrid quantum programs in PennyLane

10K 238 90

spexial

Special functions in Jax

5K 20 6

hyperjet

Algorithmic differentiation with hyper-dual numbers in C++ and Python

3K 17 4

quabla

Rust-native SciML compiler and runtime with composable autodiff and CPU/CUDA/MLX backends

3K 0 0

nufftax

Pure JAX implementation of the Non-Uniform Fast Fourier Transform (NUFFT)

2K 29 3

quabla-mlx

Rust-native SciML compiler and runtime with composable autodiff and CPU/CUDA/MLX backends

1K 0 0

tangent-ad

Source-to-Source Debuggable Derivatives in Pure Python

1K 3 0

dlux

Differentiable Physical Optics Propagation in Jax

1K 80 20

mygrad

Drop-in autodiff for NumPy.

785 216 23

hypergraph

Reversed mode second order automatic differentiation for python (WIP)

656 4 0

nabla-ml

Nabla: High-Performance Scientific Computing

572 349 16

co2-eos

Differentiable CO₂ thermodynamic properties in JAX. Pure-JAX Span-Wagner equation of state — JIT-compiled, vmappable, and fully differentiable via jax.grad. Phase-aware solvers robust through the critical point. Allows GPU speedup over CoolProp for CO₂.

553 2 1

trace-opt

End-to-end Generative Optimization for AI Agents

536 763 63

optyx

Intuitive symbolic interface for constrained optimization problems. Write natural Python, get automatic gradients and solvers.

497 25 3

gradgen

Autodiff from Python to Embedded Rust

449 8 0

jaxonomy

JAX-native simulation engine for hybrid dynamical systems — block-diagram modeling, ODE/DAE solvers, events, acausal physics, and end-to-end autodiff (grad/vmap/jit).

440 20 0