Nodes/ComfyUI-ModelScope/ModelScope VLM
ComfyUI Node

ModelScope VLM

Analyze images with text using ModelScope Vision-Language Models.

By neverbiasu·Created 11 months ago·Updated 4 months ago· 3
ModelScope VLM
    • response
    model_idQwen/QVQ-72B-Preview
    image_url
    prompt
    system_prompt
    api_key
    request_timeout120
    CategoryModelScope

    Inputs (6)

    NameTypeDefaultDescription
    model_idSTRINGQwen/QVQ-72B-PreviewModelScope VLM model ID.
    image_urlSTRINGURL of the image to analyze. Must be publicly accessible.
    promptSTRINGQuestion or instruction about the image.
    system_promptoptSTRINGOptional system prompt to guide the assistant's behavior.
    api_keyoptSTRINGModelScope API key. If empty, reads MODELSCOPE_API_KEY or MODELSCOPE_ACCESS_TOKEN from environment.
    request_timeoutoptINT12030–300Request timeout in seconds. VLM inference can be slow, recommend 120+ seconds.

    Outputs (1)

    NameTypeDescription
    responseSTRINGVLM response describing or answering about the image.