BerriAI/litellm · error · OCIError

Each content item must have a string `type` field

Error message

Each content item must have a string `type` field

What it means

During OCI GENERIC message translation, a multipart content item passed the dict check but its `type` field is absent or not a string. OpenAI content parts are discriminated by a string `type` ("text", "image_url"), and OCI needs that discriminator to map to its typed content parts, so LiteLLM rejects the message with a 400 before sending. Pure input-shape validation, fail-fast.

Solutions

1. Add a string type to every content part: {"type": "text", "text": "..."} or {"type": "image_url", "image_url": {...}}
2. Check for casing/key mistakes (Type, content_type) in the code that builds messages
3. Add a pre-flight assertion/lint on message dicts before calling completion
4. If parts come from JSON deserialization, validate against a small schema (e.g. pydantic or the type guard below)

Example fix

# before
content = [{"text": "hello"}]  # no `type`

# after
content = [{"type": "text", "text": "hello"}]

def ensure_content_part_types(content: list) -> list:
    for part in content:
        if not isinstance(part.get("type"), str):
            raise ValueError(f"content part missing string `type`: {part!r}")
    return content

def is_valid_content_part(part) -> bool:
    return isinstance(part, dict) and isinstance(part.get("type"), str)

When it happens

Trigger: litellm.completion with an oci/generic.* model and a message like {"role":"user","content":[{"text":"hi"}]} (missing type) or {"type": 1, ...} (non-string type). No API call is made; the error is raised in adapt_messages_to_generic_oci_standard_content_message.

Common situations: Typos like "Type" or omitted type when hand-crafting parts; LLM-generated message JSON missing the discriminator; schemas/downstream code that strips the type key; porting from providers where type was optional.

Related errors

• Content item of type `text` must have a string `text` field
• Content type ` ` is not supported by OCI
• Each content item must be a dictionary
• Each tool call must be a dictionary
• Message `content` must be a string or list of content parts

AI-assisted analysis of BerriAI/litellm@77b7c6c40c (2026-08-18). Data as JSON: /api/errors/27e4abc2c6d9a100. Report an issue: GitHub.

Appendix: source

Thrown at litellm/llms/oci/chat/generic.py:76

def adapt_messages_to_generic_oci_standard_content_message(role: str, content: str | list) -> OCIMessage:
    """Convert a plain-text or multipart content message to OCI format."""
    new_content: Final[list[OCIContentPartUnion]] = []
    if isinstance(content, str):
        return OCIMessage(
            role=open_ai_to_generic_oci_role_map[role],
            content=[OCITextContentPart(text=content)],
            toolCalls=None,
            toolCallId=None,
        )

    for content_item in content:
        if not isinstance(content_item, dict):
            raise OCIError(status_code=400, message="Each content item must be a dictionary")

        item_type = content_item.get("type")
        if not isinstance(item_type, str):
            raise OCIError(
                status_code=400,
                message="Each content item must have a string `type` field",
            )
        if item_type not in ["text", "image_url"]:
            raise OCIError(
                status_code=400,
                message=f"Content type `{item_type}` is not supported by OCI",
            )

        if item_type == "text":
            text = content_item.get("text")
            if not isinstance(text, str):
                raise OCIError(
                    status_code=400,
                    message="Content item of type `text` must have a string `text` field",
                )
            new_content.append(OCITextContentPart(text=text))

View on GitHub (pinned to 77b7c6c40c)