dfx-cuda-linux 1.0.11

.NET 10.0 This package targets .NET 10.0. The package is compatible with this framework or higher. 
dotnet add package dfx-cuda-linux --version 1.0.11
                    


                    

                
NuGet\Install-Package dfx-cuda-linux -Version 1.0.11
This command is intended to be used within the Package Manager Console in Visual Studio, as it uses the NuGet module's version of Install-Package. 
<PackageReference Include="dfx-cuda-linux" Version="1.0.11" />
For projects that support PackageReference, copy this XML node into the project file to reference the package. 
<PackageVersion Include="dfx-cuda-linux" Version="1.0.11" />
                    


                            Directory.Packages.props
                        

                    

                
<PackageReference Include="dfx-cuda-linux" />
                    


                            Project file
For projects that support Central Package Management (CPM), copy this XML node into the solution Directory.Packages.props file to version the package. 
paket add dfx-cuda-linux --version 1.0.11
                    


                    

                
#r "nuget: dfx-cuda-linux, 1.0.11"
#r directive can be used in F# Interactive and Polyglot Notebooks. Copy this into the interactive tool or source code of the script to reference the package. 
#:package dfx-cuda-linux@1.0.11
#:package directive can be used in C# file-based apps starting in .NET 10 preview 4. Copy this into a .cs file before any lines of code to reference the package. 
#addin nuget:?package=dfx-cuda-linux&version=1.0.11
                    


                            Install as a Cake Addin
                        

                    

                
#tool nuget:?package=dfx-cuda-linux&version=1.0.11
                    


                            Install as a Cake Tool

dfx

dfx is a tensor library for F# with support for differentiable programming. It is designed for use in machine learning, probabilistic programming, optimization and other domains.

  • Tensors with forward, reverse and nested (higher-order) automatic differentiation
  • Models, optimizers and differentiable probability distributions
  • A Torch backend using LibTorch, the C++/CUDA tensor library that powers PyTorch, on CPUs and NVIDIA GPUs
  • A Reference backend written in F#, which runs anywhere .NET runs

Which package to use

Package Contents
dfx-cpu dfx with the LibTorch CPU binaries for Windows (x64), Linux (x64) and macOS (Apple Silicon)
dfx-cuda-windows dfx with the LibTorch CUDA 12.8 binaries for Windows (x64). Large download
dfx-cuda-linux dfx with the LibTorch CUDA 12.8 binaries for Linux (x64). Large download
dfx-cuda Both dfx-cuda-windows and dfx-cuda-linux
dfx-lite dfx without LibTorch binaries, for use with a LibTorch installed by other means, for example with PyTorch

The dfx-Dfx.* packages contain the individual libraries that these packages reference.

Example

open Dfx

// The gradient of sum(x^2) is 2x
let f (x: Tensor) = (x * x).sum()
let g = dfx.grad f (dfx.tensor [1.0; 2.0; 3.0])   // tensor([2., 4., 6.])

Requirements

  • .NET 10 or later
  • CUDA packages: an NVIDIA GPU with a recent driver. The CUDA runtime is included
  • macOS: the LibTorch build for Apple Silicon needs OpenMP, which you can install with brew install libomp

Credits

dfx is the continuation of DiffSharp. Its lead creator is Atılım Güneş Baydin, and it was developed with Don Syme, Barak A. Pearlmutter, Jeffrey Mark Siskind and many contributors.

If you use dfx in your work, please cite: Atılım Güneş Baydin, Barak A. Pearlmutter, Alexey Andreyevich Radul and Jeffrey Mark Siskind. Automatic Differentiation in Machine Learning: a Survey. Journal of Machine Learning Research, 18(153):1–43, 2018.

dfx is licensed under the BSD 2-Clause License.

Product Compatible and additional computed target framework versions.
.NET net10.0 is compatible.  net10.0-android was computed.  net10.0-browser was computed.  net10.0-ios was computed.  net10.0-maccatalyst was computed.  net10.0-macos was computed.  net10.0-tvos was computed.  net10.0-windows was computed. 
Compatible target framework(s)
Included target framework(s) (in package)
Learn more about Target Frameworks and .NET Standard.

NuGet packages (1)

Showing the top 1 NuGet packages that depend on dfx-cuda-linux:

Package Downloads
dfx-cuda

dfx, a tensor library for F# with support for differentiable programming, with the LibTorch CUDA 12.8 binaries for both Windows (x64) and Linux (x64). Large download.

GitHub repositories

This package is not used by any popular GitHub repositories.

Version Downloads Last Updated
1.0.11 44 10/30/2026
1.0.10 116 10/11/2026