Specified scale type 'Standard' of account deployment is not supported by GPT4 or GPT35TURBO about openai HOT 2 OPEN

I hit the same thing. I got around it by changing the accounts/deployments bicep (in infra/core/ai/cognitiveservices.bicep) to use a newer version of the api, and setting sku instead of scaleSettings. Not 100% sure this is right, but I got this by manually deploying a model and then generating bicep from it. (I hardcoded values while I play with things but they should probably be parameters.)

@batchSize(1)
resource deployment 'Microsoft.CognitiveServices/accounts/deployments@2023-05-01' = [for deployment in deployments: {
  parent: account
  name: deployment.name
  sku: {
    name: 'Standard'
    capacity: 40
  }
  properties: {
    model: deployment.model
    raiPolicyName: contains(deployment, 'raiPolicyName') ? deployment.raiPolicyName : null
    // scaleSettings: deployment.scaleSettings
  }
}]

from openai.

I hit the same thing. I got around it by changing the accounts/deployments bicep (in infra/core/ai/cognitiveservices.bicep) to use a newer version of the api, and setting sku instead of scaleSettings. Not 100% sure this is right, but I got this by manually deploying a model and then generating bicep from it. (I hardcoded values while I play with things but they should probably be parameters.)

@batchSize(1)
resource deployment 'Microsoft.CognitiveServices/accounts/deployments@2023-05-01' = [for deployment in deployments: {
  parent: account
  name: deployment.name
  sku: {
    name: 'Standard'
    capacity: 40
  }
  properties: {
    model: deployment.model
    raiPolicyName: contains(deployment, 'raiPolicyName') ? deployment.raiPolicyName : null
    // scaleSettings: deployment.scaleSettings
  }
}]

Thank you! That worked like a charm!

from openai.