{
    "model": {
        "name": "Wan2.1 Vista4D 384p 14B",
        "architecture": "vista4d",
        "description": "Vista4D is a video reshooting framework which synthesizes the dynamic scene represented by an input source video from novel camera trajectories and viewpoints. For best efficiency, you must list all the dynamic objects / people in the 'Dynamic object keywords' input. ",
        "URLs": [
            "https://huggingface.co/DeepBeepMeep/Wan2.1/resolve/main/wan2.1_vista4d_384p49_14B_bf16.safetensors",
            "https://huggingface.co/DeepBeepMeep/Wan2.1/resolve/main/wan2.1_vista4d_384p49_14B_quanto_bf16_int8.safetensors"
        ]
    },
    "prompt": "a cinematic video of the scene with realistic motion and consistent geometry",
    "negative_prompt": "",
    "video_prompt_type": "UV",
    "resolution": "672x384",
    "fps": "control",
    "video_length": 49,
    "model_mode": "dolly_zoom",
    "num_inference_steps": 50,
    "guidance_scale": 5,
    "flow_shift": 5,
    "custom_settings": {
        "vista4d_scene_scale": 1.0,
        "vista4d_camera_strength": 100.0,
        "vista4d_seg_keywords": "_all_"
    }
}