vae/config.json

{
  "_class_name": "AutoencoderKLCogVideoX",
  "_diffusers_version": "0.31.0",
  "_name_or_path": "THUDM/CogVideoX-5b",
  "act_fn": "silu",
  "block_out_channels": [
    128,
    256,
    256,
    512
  ],
  "down_block_types": [
    "CogVideoXDownBlock3D",
    "CogVideoXDownBlock3D",
    "CogVideoXDownBlock3D",
    "CogVideoXDownBlock3D"
  ],
  "force_upcast": true,
  "in_channels": 3,
  "latent_channels": 16,
  "latents_mean": null,
  "latents_std": null,
  "layers_per_block": 3,
  "norm_eps": 1e-06,
  "norm_num_groups": 32,
  "out_channels": 3,
  "quantization_config": {
    "_load_in_4bit": true,
    "_load_in_8bit": false,
    "bnb_4bit_compute_dtype": "float16",
    "bnb_4bit_quant_storage": "uint8",
    "bnb_4bit_quant_type": "nf4",
    "bnb_4bit_use_double_quant": false,
    "llm_int8_enable_fp32_cpu_offload": false,
    "llm_int8_has_fp16_weight": false,
    "llm_int8_skip_modules": null,
    "llm_int8_threshold": 6.0,
    "load_in_4bit": true,
    "load_in_8bit": false,
    "quant_method": "bitsandbytes"
  },
  "sample_height": 480,
  "sample_width": 720,
  "scaling_factor": 0.7,
  "shift_factor": null,
  "temporal_compression_ratio": 4,
  "up_block_types": [
    "CogVideoXUpBlock3D",
    "CogVideoXUpBlock3D",
    "CogVideoXUpBlock3D",
    "CogVideoXUpBlock3D"
  ],
  "use_post_quant_conv": false,
  "use_quant_conv": false
}