<!--
{
  "availability" : [

  ],
  "documentType" : "symbol",
  "framework" : "MLCompute",
  "identifier" : "/documentation/MLCompute",
  "metadataVersion" : "0.1.0",
  "role" : "Framework",
  "symbol" : {
    "kind" : "Framework",
    "modules" : [
      "ML Compute"
    ],
    "preciseIdentifier" : "MLCompute"
  },
  "title" : "ML Compute"
}
-->

# ML Compute

Accelerate training and validation of neural networks across the CPU and one or more GPUs.

## Overview

ML Compute uses high performance <doc://com.apple.documentation/documentation/Accelerate/BNNS> primitives from the Accelerate framework for the CPU, and <doc://com.apple.documentation/documentation/MetalPerformanceShaders> for the GPU.

## Topics

### Components

[`MLCTensor`](/documentation/MLCompute/MLCTensor)

The data object you use throughout the framework.

[`MLCPlatform`](/documentation/MLCompute/MLCPlatform)

A utility class for setting global properties in the framework.

[Layers](/documentation/MLCompute/layers)

Create and inspect layers that encapsulate operations and configuration details to receive, process, and output tensors.

[Training and Validation](/documentation/MLCompute/training-and-validation)

Create, train, and validate a graph to produce acceptable prediction results.


---

Copyright &copy; 2026 Apple Inc. All rights reserved. | [Terms of Use](https://www.apple.com/legal/internet-services/terms/site.html) | [Privacy Policy](https://www.apple.com/privacy/privacy-policy)