diffusion_decoder/config.json

{
  "_class_name": "LTX2VideoDiffusionDecoderModel",
  "_diffusers_version": "0.40.0.dev0",
  "decoder_head_dim": 64,
  "decoder_model_output_type": "x0",
  "decoder_num_inference_steps": 1,
  "decoder_stage5_kernel": [
    11,
    11,
    11
  ],
  "decoder_stage_channels": [
    2048,
    1024,
    512,
    512,
    256
  ],
  "decoder_stage_depths": [
    4,
    6,
    4,
    2,
    8
  ],
  "decoder_stage_kernels": [
    [
      3,
      7,
      7
    ],
    [
      3,
      7,
      7
    ],
    [
      3,
      5,
      5
    ],
    [
      3,
      5,
      5
    ]
  ],
  "decoder_t_emb_dim": 384,
  "decoder_timestep_scale_multiplier": 1000.0,
  "decoder_upsample_channel_reductions": [
    2,
    2,
    1,
    2
  ],
  "decoder_upsample_strides": [
    [
      1,
      2,
      2
    ],
    [
      2,
      1,
      1
    ],
    [
      2,
      2,
      2
    ],
    [
      2,
      2,
      2
    ]
  ],
  "latent_channels": 128,
  "out_channels": 3,
  "patch_size": 4,
  "scaling_factor": 1.0,
  "spatial_compression_ratio": 32,
  "temporal_compression_ratio": 8
}