What causes "messages: missing text in content item" in oobabooga/textgen?

Include the text key: {"type": "text", "text": "hello"}.

oobabooga/textgen · error · InvalidRequestError

messages: missing text in content item

Error message

messages: missing text in content item

What it means

A content item with type 'text' must include a 'text' key holding the string. Its absence raises InvalidRequestError (400, param='messages') during multimodal content validation.

Source

Thrown at modules/api/completions.py:519

        # Handle multimodal content validation
        content = m.get('content')
        if content is None:
            # OpenAI allows content: null on assistant messages when tool_calls is present
            if m['role'] == 'assistant' and m.get('tool_calls'):
                m['content'] = ''
            else:
                raise InvalidRequestError(message="messages: missing content", param='messages')

        # Validate multimodal content structure
        if isinstance(content, list):
            for item in content:
                if not isinstance(item, dict) or 'type' not in item:
                    raise InvalidRequestError(message="messages: invalid content item format", param='messages')
                if item['type'] not in ['text', 'image_url']:
                    raise InvalidRequestError(message="messages: unsupported content type", param='messages')
                if item['type'] == 'text' and 'text' not in item:
                    raise InvalidRequestError(message="messages: missing text in content item", param='messages')
                if item['type'] == 'image_url' and ('image_url' not in item or 'url' not in item['image_url']):
                    raise InvalidRequestError(message="messages: missing image_url in content item", param='messages')

    # Chat Completions
    object_type = 'chat.completion' if not stream else 'chat.completion.chunk'
    created_time = int(time.time())
    cmpl_id = "chatcmpl-%d" % (int(time.time() * 1000000000))
    resp_list = 'data' if is_legacy else 'choices'

    # generation parameters
    generate_params = process_parameters(body, is_legacy=is_legacy)
    if stop_event is not None:
        generate_params['stop_event'] = stop_event
    continue_ = body['continue_']

    # Instruction template
    if body['instruction_template_str']:
        instruction_template_str = body['instruction_template_str']

View on GitHub (pinned to ed888c71f2)

Solutions

Include the text key: {"type": "text", "text": "hello"}.
If the field may be empty, still send an empty string rather than omitting the key.
Use a client-side content-item builder that always sets both keys.

Example fix

# before
{"type": "text", "content": "describe this image"}

# after
{"type": "text", "text": "describe this image"}

Defensive patterns

Strategy: validation

Validate before calling

def text_item(text: str) -> dict:
    return {'type': 'text', 'text': text or ''}

Type guard

def is_valid_text_item(i) -> bool:
    return isinstance(i, dict) and i.get('type') == 'text' and isinstance(i.get('text'), str)

Try / catch

try:
    resp = client.chat.completions.create(model=m, messages=msgs)
except openai.BadRequestError as e:
    if 'missing text in content item' in str(e):
        raise ValueError('content builder omitted text field')
    raise

Prevention

Never hand-write {'type': 'text'} inline; use a builder that sets both keys.
Use 'text', not 'content' or 'value', as the payload key.
Allow empty strings but never omit the key.

When it happens

Trigger: POST /v1/chat/completions with {"type": "text"} or {"type": "text", "content": "hi"} (wrong key name) inside a content list.

Common situations: Using 'content' or 'value' instead of 'text' as the key; constructing items from templates where the text field is conditionally omitted; partial conversion from another API's block schema.

Related errors

AI-assisted analysis of oobabooga/textgen@ed888c71f2 (2026-08-15). Data as JSON: /api/errors/c8def54ee54c7333. Report an issue: GitHub.