Skip to main content

ArcGIS Pro 3.7 system requirements

Before installing or upgrading ArcGIS Pro on a virtual or physical machine, ensure that your system meets the minimum requirements to run it. Also, learn what resources are recommended to get the best performance.

Scan your computer for compatibility

In addition to comparing your machine's compatibility with the requirements below, you can download an app that will scan your machine and generate a compatibility report: Verify that your computer can run ArcGIS Pro.

Supported operating systems

The software is supported on the Microsoft Windows operating systems listed below, as long as the operating system is also currently supported by its provider for general or public use.

Operating system Latest update or service pack tested
Windows 11 Home, Pro, and Enterprise (64 bit) April 2026 update
Windows Server 2025 Standard and Datacenter (64 bit) April 2026 update
Windows Server 2022 Standard and Datacenter (64 bit) April 2026 update
Windows Server 2019 Standard and Datacenter (64 bit) April 2026 update
Note:

  • Prior and future updates or service packs for these operating system versions are supported unless otherwise stated. The operating system version and updates must also be supported by the operating system provider.

  • The Desktop Experience option is required on all versions of Windows Server.

  • The Microsoft Media Feature Pack for Windows11 N is required to use presentations.

Software requirements

The following software must be installed separately before installing ArcGIS Pro:

Software

Minimum requirement

Microsoft .NET Desktop Runtime

Microsoft .NET Desktop Runtime 10.0.0 or a later patch release (10.0.1 and so on), using a Windows x64 installer, is required. Microsoft .NET Desktop Runtime is updated frequently. It is recommended that you install the latest patches as they become available. However, installing .NET 11 or any later version doesn't fulfill this software requirement.

WebView2 Runtime

Microsoft Edge WebView2 Runtime version 132 or later is required. If it is not already installed on the machine, use the Evergreen Bootstrapper or Evergreen Standalone Installer from Microsoft.

Note:

A per-machine installation (the default) of ArcGIS Pro requires a per-machine installation of WebView2 Runtime. To do this, right-click the WebView2 installer and click Run as administrator.

Hardware requirements

The minimum, recommended, and optimal requirements to run ArcGIS Pro are listed below. Minimum requirements will support the application, but may not perform well. Recommended requirements perform well in most situations. Best performance is gained using the optimal requirements.

Item

Supported and recommended

CPU

Minimum: 2 cores, simultaneous multithreading

Simultaneous multithreading, or hyperthreading, of CPUs typically features 2 threads per core. A multithreaded 2-core CPU will have 4 threads available for processing, while a multithreaded 6-core CPU will have 12 threads available for processing.

Full Motion Video has higher minimum and recommended CPU specifications. See Introduction to full-motion video for details.

Recommended: 4 cores

Optimal: 10 cores

Platform

x64

Storage

Minimum: 32 GB of free space

Recommended: 32 GB or more of free space on a solid-state drive (SSD)

Memory/RAM

Minimum: 8 GB

Recommended: 32 GB

Optimal: 64 GB or more

GPU

Minimum: A supported CPU (see above)

Using a configuration without any GPU leads to the emulation of one, resulting in suboptimal performance for certain visualization workflows, including 3D views.

Recommended: A discrete (not integrated) GPU that is supported by the manufacturer

This provides better performance in workflows involving visualization. For more information, see the section below on General-purpose computing on a GPU.

Dedicated (not shared) graphics memory

Recommended: 8 GB or more

If you're using a notebook computer with an integrated GPU, consider increasing the system RAM to compensate for the use of shared memory.

Visualization cache

The temporary visualization cache can consume up to 32 GB of space, if available, in the user-selected location. By default, the visualization cache is written to the user profile’s \Local subfolder, so it does not roam with the user profile if roaming profiles are enabled by your system administrator.

DirectX*

Minimum: DirectX 11, feature level 11.0, Shader Model 5.0

Recommended: DirectX 12, feature level 12.0, Shader Model 6.0

OpenGL*

Minimum: OpenGL 4.3 with the ARB_clip_control and EXT_texture_compression_s3tc extensions

Recommended: OpenGL 4.5 with the ARB_shader_draw_parameters, EXT_swap_control, EXT_texture_compression_s3tc, and EXT_texture_filter_anisotropic extensions

Vulkan*

Minimum: Version 1.3 with VK_EXT_depth_clip_enable, VK_KHR_surface, VK_KHR_swapchain, and VK_KHR_win32_surface extensions, and depthBiasClamp, depthClamp, dynamicRendering, fillModeNonSolid, geometryShader, multiViewport, samplerAnisotropy, shaderClipDistance, synchronization2, tessellationShader, and timelineSemaphore features.

Recommended: Above requirements, plus support for VK_EXT_swapchain_colorspace and VK_EXT_transform_feedback extensions.

Screen resolution

Minimum: 1024x768

Recommended: 1080p or higher

Higher resolutions, such as 4K, require more video memory and a more powerful GPU.

* At least one of the supported APIs is required. If DirectX12 requirements are not met, Vulkan, DirectX11, or OpenGL is automatically used. You can also manually change to another rendering engine, if needed, such as when you suspect an issue with your DirectX driver. If no supported GPU is present, ArcGIS Pro will use software emulation, which will result in significant performance disadvantages.
Note:

General-purpose computing on a GPU

Support for general-purpose computing on a GPU (GPGPU) using CUDA is not required to run the software but is required to run most deep learning functionality. GPGPU can also accelerate some Spatial Analyst tools.

Item

Supported and recommended

GPU type

NVIDIA GPU with CUDA compute capability 5.0 minimum; 7.5 or later recommended. See the list of CUDA-enabled cards to determine compute capability of a GPU, or check the CUDA Compute section of the system requirements checker.

GPU driver

NVIDIA GPU drivers: version 551.61 or later is required.

Dedicated graphics memory

Minimum: 6 GB

Recommended: 16 GB or more

Memory requirements depend on the model architecture and batch size being used.
Note:

  • An out-of-date GPU driver will cause deep learning tools to fail with run time errors indicating that CUDA is not installed or an unsupported tool chain is present. Verify that you have up-to-date GPU drivers directly provided by NVIDIA.

  • If you have two GPUs, you can dedicate one to visualization tasks and the other to GPGPU/CUDA tasks.

Licensing

Two license types are available for ArcGIS Pro 3.7. Depending on the type you use, you may need additional software or access to ArcGIS Online.

Note:

The Concurrent Use license type was deprecated on July 1, 2025. See the Concurrent Use License Type and ArcGIS License Manager deprecation notice for more information.

Python

See ArcGIS Pro Python environment for more information on the Python environment for ArcGIS Pro 3.7.

SDK requirements

The minimum requirements to create add-ins using ArcGIS Pro SDK are described in the SDK GitHub repository.

Virtualization

ArcGIS Pro can run in many on-premises and cloud virtual desktop environments, and requires GPU acceleration. In a properly configured, GPU-enabled environment, ArcGIS Pro behaves much like a local workstation, allowing smooth map navigation, visualization, and analysis. GPU support is also necessary when working with maps, scenes, or large datasets.

Virtualization allows ArcGIS Pro to be accessed from shared infrastructure or remote systems, but performance depends on how the environment is configured and sized for your workload.

Additional setup guidance is available in the Virtualization section of the ArcGIS Pro help documentation.

On-premises virtual desktop infrastructure (VDI)

Platform/Product

Supported environment/GPU option

Performance notes

VMware vSphere + NVIDIA vGPU

vSphere/ESXi 8.x with supported NVIDIA GPUs

Good balance of performance and scalability for shared GIS desktops.

Citrix Virtual Apps and Desktops

GPU-enabled virtual desktops

Suitable for full desktop delivery of ArcGIS Pro.

Publishing ArcGIS Pro as an app is not supported.

Microsoft Hyper-V Server

GPU pass-through or partitioning

Provides strong performance when a dedicated GPU is assigned.

Additional VDI guidance:

  • GPU acceleration greatly improves map navigation, rendering, and 3D performance.

  • Avoid environments that do not provide GPU support.

  • Application virtualization or layering technologies may work, but performance and reliability can be limited.

  • Match system sizing to workload—Heavy 3D and analysis workflows require more resources.

Cloud infrastructure

Cloud platforms allow ArcGIS Pro to run on GPU-enabled virtual machines or managed desktops. Choosing the right instance type ensures smooth navigation and analysis performance.

Most GPU-enabled cloud platforms support scalable instance sizing, allowing you to start with a smaller configuration and increase CPU, memory, or GPU resources as workload demands grow.

Scaling primarily affects how smoothly large datasets render, how quickly scenes load, and how responsive analysis workflows feel.

Cloud platform/product

Instance type/GPU option

Performance notes

AWS EC2

G4dn instance family

Scalable GPU desktops suitable for everyday mapping and moderate GIS workflows.

AWS EC2

G5 instance family

Higher-performance, scalable GPU desktops for advanced visualization and analysis.

AWS EC2

G6/G6f instance families

Scalable, next-generation GPU options designed for complex 3D scenes and large datasets.

AWS WorkSpaces

Managed GPU desktops

Managed desktops that can be sized for different GIS workloads.

Google Cloud Platform Compute Engine

GPU-enabled VM instances (NVIDIA T4, L4)

Scalable GPU desktops suitable for ArcGIS Pro visualization and analysis. Instance sizing can be increased for more demanding GID workflows.

Microsoft Azure virtual machines

NVads_A10 v5 series

Scalable, GPU-enabled desktops supporting 2D and 3D GIS workflows.

Microsoft Azure virtual machines

NCasT4_v3 series

GPU-enabled, scalable machines suitable for compute and visualization workloads.

Microsoft Azure Cloud PCs (Windows 365)

Managed cloud desktops with GPU-enabled options

Cloud desktops that can be configured to match GIS workload requirements.

Additional cloud guidance:

  • Most cloud GPU environments allow scaling as workloads grow.

  • AWS EC2 example: start with a smaller G4dn instance and scale to G5 for heavier work.

  • Azure VM example: start with smaller NVads_A10 sizes and scale as needed.

  • Larger datasets and 3D scenes benefit from higher GPU capacity.

  • Increasing instance size typically improves responsiveness.

  • GPU-enabled environments support smoother visualization and analysis workflows.

Cloud application streaming and virtualization

Some platforms allow ArcGIS Pro to be delivered through streamed sessions or virtual desktops instead of a traditional workstation install. While application streaming is possible, a full desktop presentation provides the best ArcGIS Pro experience.

Full desktop environments offer better GPU access, system resource availability, and session stability, resulting in smoother navigation and more predictable performance.

Platform/Product

Deployment model

Performance notes

Microsoft Azure Virtual Desktop (AVD)

GPU-enabled, full virtual desktops

Recommended approach for delivering ArcGIS Pro in the cloud.

Amazon AppStream 2.0

Application or desktop streaming

Works for many workflows, but full desktop delivery typically provides better responsiveness.

What's new

See What's new in ArcGIS Pro 3.7 to learn about new functionality and improvements made since the previous software release.