Azure OpenAI content filter response breaks streamText

[jb-dev1]

Description

When a user prompt is blocked using the AI SDK's Google Generative AI provider with "HARM_CATEGORY_DANGEROUS_CONTENT" (as an example), the onFinish finishReason = content-filter which allows me to capture the value and move on. However, when a user prompt is filtered using the AI SDK's Azure OpenAI provider where the model endpoint's content filter is triggered, the errors below are encountered.

My expected result is an onFinish finishReason = content-filter.

"@ai-sdk/azure": "^1.0.13",
"@ai-sdk/google": "^1.0.12",
"ai": "4.0.10",

 ⨯ Error: failed to pipe response
    at pipeToNodeResponse

[cause]: [Error [AI_APICallError]: The response was filtered due to the prompt triggering Azure OpenAI's content management policy. Please modify your prompt and retry. To learn more about our content filtering policies please read our documentation: https://go.microsoft.com/fwlink/?linkid=2198766] {
    cause: undefined,
    url: 'https://<hidden>.openai.azure.com/openai/deployments/gpt-4o-mini-safety-test/chat/completions?api-version=2024-10-01-preview',
    requestBodyValues: {
      model: 'gpt-4o-mini',
      temperature: 0.6,
      messages: [Array],
      stream: true,
      stream_options: [Object]
      <some values hidden>
    },
    statusCode: 400,
    responseHeaders: {
      'apim-request-id': 'd21fc18c-8a1f-49a7-927b-3cece6bbf39d',
      'azureml-model-session': 'd090-20241221033500',
      'content-length': '666',
      'content-type': 'application/json',
      date: 'Sun, 29 Dec 2024 15:47:13 GMT',
      'ms-azureml-model-error-reason': 'model_error',
      'ms-azureml-model-error-statuscode': '400',
      'strict-transport-security': 'max-age=31536000; includeSubDomains; preload',
      'x-content-type-options': 'nosniff',
      'x-envoy-upstream-service-time': '169',
      'x-ms-client-request-id': 'd21fc18c-8a1f-49a7-927b-3cece6bbf39d',
      'x-ms-rai-invoked': 'true',
      'x-ms-region': 'East US',
      'x-ratelimit-remaining-requests': '98',
      'x-ratelimit-remaining-tokens': '8674',
      'x-request-id': 'd076474e-fda0-4363-97cc-4d43e1446ff6'
    },
    responseBody: `{"error":{"message":"The response was filtered due to the prompt triggering Azure OpenAI's content management policy. Please modify your prompt and retry. To learn more about our content filtering policies please read our documentation: https://go.microsoft.com/fwlink/?linkid=2198766","type":null,"param":"prompt","code":"content_filter","status":400,"innererror":{"code":"ResponsibleAIPolicyViolation","content_filter_result":{"hate":{"filtered":true,"severity":"low"},"jailbreak":{"filtered":false,"detected":false},"self_harm":{"filtered":false,"severity":"safe"},"sexual":{"filtered":false,"severity":"safe"},"violence":{"filtered":true,"severity":"medium"}}}}}`,
    isRetryable: false,
    data: { error: [Object] }
  }
}

 POST /api/chat 500 in 1019ms
 ⨯ ../../src/server/pipe-readable.ts (144:11) @ pipeToNodeResponse
 ⨯ Internal error: Error: failed to pipe response

Code example

const result = streamText({
  model: azure("gpt-4o-mini-safety-test"),
  system: systemPrompt,
  messages: coreMessages,
  temperature: 0.7,
  onFinish: async ({ finishReason, usage, warnings, response }) => {
    console.log("onFinish Callback Triggered:");
    console.log("Finish Reason:", finishReason);
    console.log("Token Usage:", usage);
    console.log("Warnings (if any):", warnings);
    console.log("Response:", response);

AI provider

@ai-sdk/azure 1.0.13

Additional context

No response

[jb-dev1]

May be related to #3655

Updated to "ai": "4.0.20", with no improvement.