microsoft/semantic-kernel · error · ValueError

Model ID does not contain a valid model provider name.

Error message

Model ID {model_id} does not contain a valid model provider name.

What it means

Raised by BedrockModelProvider.to_model_provider when none of the known provider substrings ('ai21', 'amazon', 'anthropic', 'cohere', 'meta', 'mistral') appear in the supplied model_id. The connector routes requests to provider-specific builders by matching the provider name inside the model ID string, so an unrecognized ID cannot be routed.

Solutions

1. Use the full Bedrock model ID including the provider prefix, e.g. 'anthropic.claude-3-5-sonnet-20240620-v1:0', 'amazon.titan-embed-text-v2:0', 'meta.llama3-8b-instruct-v1:0'.
2. If using a cross-region or provisioned inference profile ARN, resolve the underlying base model ID first, or pass an explicit model_provider argument to bypass string matching.
3. If the provider is genuinely new (not in the BedrockModelProvider enum), request a library update; do not edit the enum in a fork unless you own the build.
4. Validate the model_id against your Bedrock account's list of available foundation models before constructing the service.

Example fix

// before
service = BedrockChatCompletion(model_id="claude-3-sonnet")
// after
service = BedrockChatCompletion(model_id="anthropic.claude-3-5-sonnet-20240620-v1:0")

SUPPORTED_PROVIDERS = ("ai21", "amazon", "anthropic", "cohere", "meta", "mistral")

def has_valid_bedrock_provider(model_id: str) -> bool:
    return any(p in model_id for p in SUPPORTED_PROVIDERS)

SUPPORTED_PROVIDERS = ("ai21", "amazon", "anthropic", "cohere", "meta", "mistral")

def is_valid_bedrock_model_id(model_id: object) -> bool:
    return isinstance(model_id, str) and any(p in model_id for p in SUPPORTED_PROVIDERS)

try:
    service = BedrockChatCompletion(model_id=model_id)
except ValueError as e:
    if "does not contain a valid model provider name" in str(e):
        raise ValueError(f"Use a full Bedrock model ID like 'anthropic.claude-3-5-sonnet-...'. Got: {model_id}") from e
    raise

When it happens

Trigger: Any Bedrock text/chat/embedding call that goes through get_*_request_body or parse_*_response without an explicit model_provider argument. Fires for a model_id like 'my-custom-model', 'llama-3-8b' (missing 'meta'), or an ARN/provisioned-model ID that does not contain one of the six provider keywords.

Common situations: User passes a short model alias instead of the full Bedrock model ID (e.g. 'claude-3' instead of 'anthropic.claude-3-sonnet'); using a new provider Bedrock added that SK has not mapped yet; passing an inference-profile ARN; copy-paste typo in the model ID.

Related errors

• The Amazon Bedrock Chat Model ID is missing.
• The Amazon Bedrock Text Embedding Model ID is missing.
• The Amazon Bedrock Text Model ID is missing.
• Failed to initialize the Amazon Bedrock Agent settings
• Failed to initialize the Amazon Bedrock Chat Completion…

AI-assisted analysis of microsoft/semantic-kernel@c028a0c7dc (2026-08-13). Data as JSON: /api/errors/ea8245fca8df7dbf. Report an issue: GitHub.

Appendix: source

Thrown at python/semantic_kernel/connectors/ai/bedrock/services/model_provider/bedrock_model_provider.py:43

    """Amazon Bedrock Model Provider Enum.

    This list contains the providers of all base models available on Amazon Bedrock.
    """

    AI21LABS = "ai21"
    AMAZON = "amazon"
    ANTHROPIC = "anthropic"
    COHERE = "cohere"
    META = "meta"
    MISTRALAI = "mistral"

    @classmethod
    def to_model_provider(cls, model_id: str) -> "BedrockModelProvider":
        """Convert a model ID to a model provider."""
        try:
            return next(provider for provider in cls if provider.value in model_id)
        except StopIteration:
            raise ValueError(f"Model ID {model_id} does not contain a valid model provider name.")


# region Text Completion


TEXT_COMPLETION_REQUEST_BODY_MAPPING: dict[
    BedrockModelProvider, Callable[[str, BedrockTextPromptExecutionSettings], Any]
] = {
    BedrockModelProvider.AMAZON: bedrock_amazon_titan.get_text_completion_request_body,
    BedrockModelProvider.ANTHROPIC: bedrock_anthropic_claude.get_text_completion_request_body,
    BedrockModelProvider.COHERE: bedrock_cohere.get_text_completion_request_body,
    BedrockModelProvider.AI21LABS: bedrock_ai21_labs.get_text_completion_request_body,
    BedrockModelProvider.META: bedrock_meta_llama.get_text_completion_request_body,
    BedrockModelProvider.MISTRALAI: bedrock_mistralai.get_text_completion_request_body,
}

TEXT_COMPLETION_RESPONSE_MAPPING: dict[BedrockModelProvider, Callable[[dict[str, Any], str], list[TextContent]]] = {
    BedrockModelProvider.AMAZON: bedrock_amazon_titan.parse_text_completion_response,

View on GitHub (pinned to c028a0c7dc)