🎬 Loading...
    Prompt

    1. Target duration: 10 seconds 2. Format: 16:9 style_and_tone: ASTROCINEMAV01K2T. Photorealistic neo-noir crime drama, grounded contemporary feature-film photography, wet nighttime streets, motivated practical lighting, sodium-vapor warmth mixed with cold storefront spill, deep but detailed shadows, restrained color grading, shallow depth of field, subtle halation around practical lights, realistic skin texture, organic fine film grain. audio_design: Low restrained neo-noir score with sparse piano, distant upright bass and a barely audible dark synth layer. Continuous rainfall, dripping gutters, distant city traffic, occasional tires passing over wet asphalt and a buzzing neon transformer. Dialogue close, natural and clearly intelligible. continuous_shot: [Continuous Shot] (00:00.000–10.000) A tired middle-aged detective in a dark wool overcoat stands beneath a failing neon sign in a narrow rain-soaked alley at night. The camera begins in a medium-wide frontal composition with foreground rain catching the light and slowly dollies toward him. He removes a folded photograph from his coat pocket, studies it, then raises his eyes toward the darkness at the far end of the alley. At 00:04.000 he says quietly: "Three nights... and I'm still chasing a ghost." A passing car briefly throws moving white light across his face. He folds the photograph and turns toward a faint metallic sound deeper in the alley. At 00:08.000 another distant crash echoes. He mutters: "There you are." The camera finishes in a tight medium close-up as the score resolves on a dark sustained note.

    Generation Settings

    Parameters used to generate this content

    Created with
    ComfyUI
    Sampler
    res_multistep
    Seed173949271573699
    Steps20
    ComfyUI Workflow

    The node graph used to generate this content — pan and zoom to explore, or download it to run in ComfyUI.

    Info
    Video
    Likes
    28
    Created
    8/11/2026
    Base Model
    MiniMax H3
    Source
    CivitAI
    Models Used

    AI models used to generate this content

    Actions