oobabooga/textgen · error · InvalidRequestError

messages: missing image_url in content item

Error message

messages: missing image_url in content item

What it means

A content item with type 'image_url' must contain an 'image_url' object which itself contains a 'url' key. Missing either raises InvalidRequestError (400, param='messages'). The url may be an http(s) link or a base64 data URI.

Source

Thrown at modules/api/completions.py:521

        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']
    elif body['instruction_template']:
        instruction_template = body['instruction_template']

View on GitHub (pinned to ed888c71f2)

Solutions

1. Use the full nesting: {"type": "image_url", "image_url": {"url": "https://..." or "data:image/jpeg;base64,..."}}.
2. Check the URL is non-empty and a valid string before appending the item.
3. For local files, base64-encode into a data URI since the server fetches URLs server-side.

Example fix

# before
{"type": "image_url", "url": "https://example.com/cat.png"}

# after
{"type": "image_url", "image_url": {"url": "https://example.com/cat.png"}}

def image_item(url: str) -> dict:
    if not url or not isinstance(url, str):
        raise ValueError('image url required')
    return {'type': 'image_url', 'image_url': {'url': url}}

def is_valid_image_item(i) -> bool:
    return (isinstance(i, dict) and i.get('type') == 'image_url'
            and isinstance(i.get('image_url'), dict)
            and isinstance(i['image_url'].get('url'), str) and bool(i['image_url']['url']))

try:
    resp = client.chat.completions.create(model=m, messages=msgs)
except openai.BadRequestError as e:
    if 'missing image_url' in str(e):
        raise ValueError('image content item missing nested image_url.url')
    raise

When it happens

Trigger: POST /v1/chat/completions with {"type": "image_url"} alone, {"type": "image_url", "url": u} (url at wrong level), or {"type": "image_url", "image_url": {}} (no url).

Common situations: Flattening the nested structure by mistake; sending {'image_url': {'url': None}} from a pipeline where image loading failed; adapting from a schema that puts the URL at item top level.

Related errors

• messages: invalid content item format
• messages: unsupported content type
• messages: missing text in content item
• functions is not supported.
• function_call is not supported.

AI-assisted analysis of oobabooga/textgen@ed888c71f2 (2026-08-15). Data as JSON: /api/errors/d4c370e9a5ab3f84. Report an issue: GitHub.