<!--
{
  "availability" : [
    "iOS: 26.0.0 -",
    "iPadOS: 26.0.0 -",
    "macCatalyst: 26.0.0 -",
    "macOS: 26.0.0 -",
    "tvOS: 26.0.0 -",
    "visionOS: 26.0.0 -"
  ],
  "documentType" : "symbol",
  "framework" : "Metal",
  "identifier" : "/documentation/Metal/MTL4TileRenderPipelineDescriptor/requiredThreadsPerThreadgroup",
  "metadataVersion" : "0.1.0",
  "role" : "Instance Property",
  "symbol" : {
    "kind" : "Instance Property",
    "modules" : [
      "Metal"
    ],
    "preciseIdentifier" : "c:objc(cs)MTL4TileRenderPipelineDescriptor(py)requiredThreadsPerThreadgroup"
  },
  "title" : "requiredThreadsPerThreadgroup"
}
-->

# requiredThreadsPerThreadgroup

Sets the required number of threads per threadgroup for tile dispatches.

```
var requiredThreadsPerThreadgroup: MTLSize { get set }
```

## Discussion

This value is typically optional, except in the cases where the tile function that [`tileFunctionDescriptor`](/documentation/Metal/MTL4TileRenderPipelineDescriptor/tileFunctionDescriptor)
references uses `CooperativeTensors`. In this case, you need to provide a non-zero value to this property.

Additionally, when you set this value, the `threadsPerTile` argument of any tile dispatch needs to match it.

Setting this value to a size of 0 in every dimension disables this property.

---

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)