Nodes/digit-comfyui/DIGIT Captioner
ComfyUI Node

DIGIT Captioner

A ComfyUI node in DIGIT with 13 inputs and 3 outputs.

By thedepartmentofexternalservices·Created 6 months ago·Updated 6 days ago· 0
DIGIT Captioner
    • report
    • last_caption
    • captioned_count
    dataset_path
    actioncaption_uncaptioned
    caption_preset
    system_promptYou are an expert image captioner for AI training datasets. Describe the image in detail, focusing on subject, composition, lighting, colors, style, and mood. Be specific and descriptive. Output only the caption, no preamble.
    prompt_templateDescribe this image in detail for AI training:
    modelgemini-2.5-flash
    temperature0.40
    max_tokens300
    overwritefalse
    caption_ext.txt
    single_image_path
    gcp_project_id
    gcp_region
    CategoryDIGIT

    Inputs (13)

    NameTypeDefaultDescription
    dataset_pathSTRING
    actionCOMBOcaption_uncaptioned5 options: caption_all, caption_uncaptioned, caption_single, recaption_all, preview
    caption_presetoptSTRING
    system_promptoptSTRINGYou are an expert image captioner for AI training datasets. Describe the image in detail, focusing on subject, composition, lighting, colors, style, and mood. Be specific and descriptive. Output only the caption, no preamble.
    prompt_templateoptSTRINGDescribe this image in detail for AI training:
    modeloptCOMBOgemini-2.5-flash4 options: gemini-2.5-flash, gemini-2.5-pro, gemini-2.0-flash, gemini-2.0-flash-lite
    temperatureoptFLOAT0.400–2
    max_tokensoptINT30050–2000
    overwriteoptBOOLEANfalse
    caption_extoptSTRING.txt
    single_image_pathoptSTRING
    gcp_project_idoptSTRING
    gcp_regionoptSTRING

    Outputs (3)

    NameTypeDescription
    reportSTRING
    last_captionSTRING
    captioned_countINT