# Comparison of Julia autodiff packages
**URL:** https://discourse.julialang.org/t/comparison-of-julia-autodiff-packages/119965
**Category:** General Usage
**Tags:** autodiff
**Created:** [September 27, 2024, 1:56pm UTC](https://discourse.julialang.org/t/comparison-of-julia-autodiff-packages/119965 "2024-09-27T13:56:22Z")
**Posts on this page:** 2
**Page:** 1
### Author:  [@Tarny\_GG\_Channie](https://discourse.julialang.org/u/Tarny_GG_Channie)
#### Post date: [September 27, 2024, 1:56pm UTC](https://discourse.julialang.org/t/comparison-of-julia-autodiff-packages/119965/1 "2024-09-27T13:56:22Z")
Julia now has so many different autodiff packages. Can somebody summarize the status of these packages?
From what I know, there is Forwarddiff (which uses dual numbers), and then there is the Reversediff, which I’m not sure about. Then there is Zygote, which is said to be quite good and general, Enzyme, which operates at the LLVM level, and Diffractor, an experimental package non-contributors are recommended to ignore for now.
Then there are more.
How are they right now?
---
### Author:  [@nilshg](https://discourse.julialang.org/u/nilshg)
#### Post date: [September 27, 2024, 2:38pm UTC](https://discourse.julialang.org/t/comparison-of-julia-autodiff-packages/119965/2 "2024-09-27T14:38:37Z")
> [@State of AD in 2024](https://discourse.julialang.org/t/state-of-ad-in-2024/112601):
>
> Hi, I’m new to the Julia language and am trying to understand the AD ecosystem. From what I understand, Julia wants to broaden AD beyond the type of models you might find in Pytorch and Jax while also retaining similar performance on CPU and GPU. I read that Zygote differentiates on low level Julia IR but now the attention is towards other projects like Diffractor and Enzyme which differentiate at an even finer level. Anyways, I found [this post](https://discourse.julialang.org/t/whats-the-state-of-automatic-differentiation-in-julia-january-2023/92473) discussing the state of AD around last year, so p…