model_index.json

{
  "_class_name": "VidXTendPipeline",
  "_diffusers_version": "0.26.1",
  "controlnet": [
    "vidxtend.models.controlnet",
    "ControlNetModel"
  ],
  "resampler": [
    "vidxtend.models.image_embedder",
    "ImageEmbeddingContextResampler"
  ],
  "scheduler": [
    "diffusers",
    "DDIMScheduler"
  ],
  "num_frames": 16,
  "num_frames_conditioning": 8,
  "temp_attend_on_uncond_include_past": false,
  "temp_attend_on_neighborhood_of_condition_frames": false,
  "temporal_self_attention_mask_included_itself": false,
  "temporal_self_attention_only_on_conditioning": false,
  "spatial_attend_on_condition_frames": false,
  "image_encoder_version": "laion2b_s32b_b79k",
  "text_encoder": [
    "transformers",
    "CLIPTextModel"
  ],
  "tokenizer": [
    "transformers",
    "CLIPTokenizer"
  ],
  "unet": [
    "vidxtend.models.unet_3d_condition",
    "UNet3DConditionModel"
  ],
  "vae": [
    "diffusers",
    "AutoencoderKL"
  ]
}