BerriAI/litellm · error · ValueError

messages parameter is required for Converse input

Error message

messages parameter is required for Converse input

What it means

Raised by validate_count_tokens_request when the request's input is detected as Converse format (it has a 'messages' key / messages-based shape) but the messages value is empty or absent. Converse-format requests are identified by structure, and token counting requires at least one message to count.

Solutions

1. Include at least one message: messages=[{'role': 'user', 'content': 'hello'}]
2. If you meant InvokeModel format, remove the 'messages' key so the detector routes to the inputText-style branch
3. Double-check that your serialization isn't dropping the field (requests library ignoring None, JSON schema stripping empty lists)

Example fix

# before
body = {'model': 'bedrock/...', 'messages': []}

# after
body = {'model': 'bedrock/...', 'messages': [{'role': 'user', 'content': 'hello'}]}

def validate_converse_messages(body: dict) -> None:
    if 'messages' in body or 'system' in body:  # converse-shaped
        msgs = body.get('messages')
        if not isinstance(msgs, list) or len(msgs) == 0:
            raise ValueError('Converse count_tokens body needs a non-empty messages list')

def is_converse_with_messages(body: dict) -> bool:
    if 'messages' not in body:
        return False
    msgs = body['messages']
    return isinstance(msgs, list) and len(msgs) > 0

try:
    handler.count_tokens(body)
except ValueError as e:
    if 'messages parameter is required' in str(e):
        raise ConversationEmptyError('nothing to count') from e
    raise

When it happens

Trigger: Sending {'model': '...', 'messages': []} or omitting messages entirely while other keys (or the detector heuristic) mark the request as Converse-style; passing messages in a different key ('prompt', 'input') so the real list looks empty.

Common situations: Building the request dynamically and filtering out all messages (e.g. empty conversation guard); confusing InvokeModel bodies ({'inputText': ...}) with Converse bodies; testing with placeholder empty payloads.

Understand the failure class

Background: "missing required argument" and "the following required arguments were not provided": what required-argument errors mean and how to fix them — this error's family across 20 libraries.

Related errors

• Message must be a dictionary
• Message must have a 'content' field
• Message must have a 'role' field
• messages must be a list
• model parameter is required

AI-assisted analysis of BerriAI/litellm@77b7c6c40c (2026-08-18). Data as JSON: /api/errors/fba3f3b32dc74384. Report an issue: GitHub.

Appendix: source

Thrown at litellm/llms/bedrock/count_tokens/transformation.py:264

        Validate the incoming count tokens request.
        Supports both Converse and InvokeModel input formats.

        Args:
            request_data: The request payload

        Raises:
            ValueError: If the request is invalid
        """
        if not request_data.get("model"):
            raise ValueError("model parameter is required")

        input_type: Final = self._detect_input_type(request_data)

        if input_type == "converse":
            # Validate Converse format (messages-based)
            messages: Final = request_data.get("messages", [])
            if not messages:
                raise ValueError("messages parameter is required for Converse input")

            if not isinstance(messages, list):
                raise ValueError("messages must be a list")

            for i, message in enumerate(messages):
                if not isinstance(message, dict):
                    raise ValueError(f"Message {i} must be a dictionary")

                if "role" not in message:
                    raise ValueError(f"Message {i} must have a 'role' field")

                if "content" not in message:
                    raise ValueError(f"Message {i} must have a 'content' field")
        else:
            # For InvokeModel format, we need at least some content to count tokens
            # The content structure varies by model, so we do minimal validation
            if len(request_data) <= 1:  # Only has 'model' field
                raise ValueError("Request must contain content to count tokens")

View on GitHub (pinned to 77b7c6c40c)