serviceTier in GenerateContentConfig is silently ignored when using Vertex AI (vertexai: true)

[gwthompson23]

When using the @google/genai SDK with vertexai: true, the serviceTier field in GenerateContentConfig has no effect. The SDK places serviceTier in the JSON request body, but the Vertex AI API does not read flex/priority tier configuration from the request body. It uses HTTP headers (X-Vertex-AI-LLM-Request-Type and X-Vertex-AI-LLM-Shared-Request-Type).

This means serviceTier: ServiceTier.FLEX silently does nothing on Vertex AI. There is no error. The request succeeds but is billed at standard rates.

Steps to reproduce:

import { GoogleGenAI, ServiceTier } from "@google/genai";

const client = new GoogleGenAI({
  vertexai: true,
  project: "my-project",
  location: "global",
});

const response = await client.models.generateContent({
  model: "gemini-3-flash-preview",
  contents: [{ role: "user", parts: [{ text: "Hello" }] }],
  config: {
    serviceTier: ServiceTier.FLEX,
  },
});

console.log(response.usageMetadata?.trafficType);
// Expected: "ON_DEMAND_FLEX"
// Actual: "ON_DEMAND"

Workaround:

Pass the Vertex AI headers manually via httpOptions:

config: {
  httpOptions: {
    headers: {
      "X-Vertex-AI-LLM-Request-Type": "shared",
      "X-Vertex-AI-LLM-Shared-Request-Type": "flex",
    },
  },
},

This correctly returns trafficType: "ON_DEMAND_FLEX".

Expected behavior:

When vertexai: true, the SDK should translate serviceTier: ServiceTier.FLEX into the appropriate X-Vertex-AI-LLM-* HTTP headers, so users don't need to know about the underlying API differences.

Additional context:

• SDK version: @google/genai@1.48.0
• The serviceTier field works correctly on the Gemini Developer API (vertexai: false)
• Flex PayGo docs confirm headers are the mechanism for Vertex AI: https://cloud.google.com/vertex-ai/generative-ai/docs/flex-paygo

[stavan2003]

With google-genai==1.72.0 python pkg I currently see a validation error being returned. Looks like its related, not very sure though

    client = Client(
        vertexai=True,
        project="project-abc,
        location="global",
    )

    response = client.models.generate_content(
        model="gemini-2.5-flash",
        contents="Triage this incoming security alert immediately.",
        config=GenerateContentConfig(
            service_tier=ServiceTier.PRIORITY,
        )
    )

Error

google.genai.errors.ClientError: 400 INVALID_ARGUMENT. {'error': {'code': 400, 'message': 'Invalid value at \'service_tier\' (type.googleapis.com/google.cloud.aiplatform.v1beta1.ServiceTier), "priority"', 'status': 'INVALID_ARGUMENT', 'details': [{'@type': 'type.googleapis.com/google.rpc.BadRequest', 'fieldViolations': [{'field': 'service_tier', 'description': 'Invalid value at \'service_tier\' (type.googleapis.com/google.cloud.aiplatform.v1beta1.ServiceTier), "priority"'}]}]}}

[Varun-S10]

Hi @gwthompson23, Thank you for bringing this to our attention. We truly appreciate you flagging this issue. We have reproduced it on our end and will be filing a bug internally.