What causes "DashScope VLM model must be explicitly selected." in ATH-MaaS/Pixelle-Video?

Pass an explicit model, e.g. query(prompt, model="qwen-vl-max")

ATH-MaaS/Pixelle-Video · error · RuntimeError

DashScope VLM model must be explicitly selected.

Error message

DashScope VLM model must be explicitly selected.

What it means

VlmClient.query requires an explicit DashScope VLM model name; it raises RuntimeError when the model argument is missing, None, or an empty/whitespace string. The client deliberately does not guess a default model.

Source

Thrown at pixelle_video/services/api_services/vlm_client.py:36

        dashscope_key = dashscope_api_key or Config.DASHSCOPE_API_KEY

        self.dashscope_client = (
            QwenVLClient(
                api_key=dashscope_key,
                base_url=dashscope_base_url or Config.DASHSCOPE_BASE_URL
            )
            if dashscope_key else None
        )

    def query(self,
             prompt: str,
             image_paths: Optional[List[str]] = None,
             model: Optional[str] = None,
             session_id: Optional[str] = None,
             video_paths: Optional[List[str]] = None) -> str:
        selected_model = (model or "").strip()
        if not selected_model:
            raise RuntimeError("DashScope VLM model must be explicitly selected.")

        if Config.PRINT_MODEL_INPUT:
            print("---- VLM REQUEST ----")
            print(f"Prompt: {prompt}")
            if image_paths:
                print(f"Images: {len(image_paths)}")
                for p in image_paths:
                    if p.startswith("data:"):
                        print(f" - [Base64图片]")
                    else:
                        print(f" - {p}")
            if video_paths:
                print(f"Videos: {len(video_paths)}")
                for p in video_paths:
                    print(f" - {p}")
            print(f"Model: {selected_model}")
            if session_id:
                print(f"Session ID: {session_id}")

View on GitHub (pinned to 848b054e4f)

Solutions

Pass an explicit model, e.g. query(prompt, model="qwen-vl-max")
Ensure your config/settings object actually has a non-empty model value before calling
If a caller passes Optional model, validate/coalesce it at that layer before invoking query

Example fix

// before
vlm.query(prompt=prompt, image_paths=[img])
// after
vlm.query(prompt=prompt, image_paths=[img], model="qwen-vl-max")

Defensive patterns

Strategy: validation

Validate before calling

def ensure_model(model: str | None) -> str:
    m = (model or "").strip()
    if not m:
        raise ValueError("DashScope VLM model must be configured (e.g. 'qwen-vl-max')")
    return m

Type guard

def has_model(model: str | None) -> bool:
    return isinstance(model, str) and bool(model.strip())

Try / catch

try:
    answer = vlm.query(prompt=p, image_paths=[img], model=cfg.vlm_model)
except RuntimeError as e:
    if "must be explicitly selected" in str(e):
        logging.critical("VLM model not configured; set it in config")
    else:
        raise

Prevention

Load the model name from config and assert it is non-empty at startup
Never forward an Optional model straight into query(); coalesce or validate first
List allowed VLM models and validate against them

When it happens

Trigger: Calling vlm_client.query(prompt, image_paths=[...]) without model=, or with model="" / model=" ".

Common situations: Caller that used to pass a default model refactored to Optional and stopped supplying it; config value for model name empty in settings; upstream code forwards its own None model parameter.

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

AI-assisted analysis of ATH-MaaS/Pixelle-Video@848b054e4f (2026-08-30). Data as JSON: /api/errors/5c88d20d9b9c46e1. Report an issue: GitHub.