microsoft/semantic-kernel · error · ServiceInitializationError

The Vertex AI embedding model ID is required.

Error message

The Vertex AI embedding model ID is required.

What it means

Raised by VertexAITextEmbedding.__init__ after settings validate but embedding_model_id is falsy. The embedding service cannot target a model without an id, so initialization aborts.

Source

Thrown at python/semantic_kernel/connectors/ai/google/vertex_ai/services/vertex_ai_text_embedding.py:70

            project_id (str): The Google Cloud project ID.
            region (str): The Google Cloud region.
            embedding_model_id (str): The Gemini model ID.
            service_id (str): The Vertex AI service ID.
            env_file_path (str): The path to the environment file.
            env_file_encoding (str): The encoding of the environment file.
        """
        try:
            vertex_ai_settings = VertexAISettings(
                project_id=project_id,
                region=region,
                embedding_model_id=embedding_model_id,
                env_file_path=env_file_path,
                env_file_encoding=env_file_encoding,
            )
        except ValidationError as e:
            raise ServiceInitializationError(f"Failed to validate Vertex AI settings: {e}") from e
        if not vertex_ai_settings.embedding_model_id:
            raise ServiceInitializationError("The Vertex AI embedding model ID is required.")

        super().__init__(
            ai_model_id=vertex_ai_settings.embedding_model_id,
            service_id=service_id or vertex_ai_settings.embedding_model_id,
            service_settings=vertex_ai_settings,
        )

    @override
    async def generate_embeddings(
        self,
        texts: list[str],
        settings: "PromptExecutionSettings | None" = None,
        **kwargs: Any,
    ) -> ndarray:
        raw_embeddings = await self.generate_raw_embeddings(texts, settings, **kwargs)
        return array(raw_embeddings)

    @override

View on GitHub (pinned to c028a0c7dc)

Solutions

1. Pass embedding_model_id explicitly (e.g. 'textembedding-gecko@003').
2. Set the embedding model id via environment variable / .env.
3. Ensure the id is non-empty and matches a deployed Vertex AI embedding model.

Example fix

# before
svc = VertexAITextEmbedding(project_id='p', region='us-central1')
# after
svc = VertexAITextEmbedding(project_id='p', region='us-central1', embedding_model_id='textembedding-gecko@003')

emb_id = os.environ.get('VERTEX_AI_EMBEDDING_MODEL_ID') or passed_embedding_model_id
assert emb_id, 'Vertex AI embedding model ID is required'

When it happens

Trigger: Constructing VertexAITextEmbedding without embedding_model_id and no corresponding env var. Passing an empty string.

Common situations: Forgetting to set the embedding model env var. Assuming a default embedding model exists. Copying sample code that omits the model argument.

Related errors

• The Vertex AI Gemini model ID is required.
• The Vertex AI Gemini model ID is required.
• Failed to validate Vertex AI settings: {e}
• Ollama embedding model ID is not set.
• The Amazon Bedrock Text Embedding Model ID is missing.

AI-assisted analysis of microsoft/semantic-kernel@c028a0c7dc (2026-08-13). Data as JSON: /api/errors/b8cbf6239afb8835. Report an issue: GitHub.