ming_image/vae_config.json

{
  "_class_name": "AutoencoderKLQwenImage",
  "_diffusers_version": "0.36.0.dev0",
  "attn_scales": [],
  "base_dim": 96,
  "dim_mult": [
    1,
    2,
    4,
    4
  ],
  "dropout": 0.0,
  "input_channels": 4,
  "latents_mean": null,
  "latents_std": null,
  "scaling_factor": 8.0064,
  "shift_factor": 0.0,
  "num_res_blocks": 2,
  "temperal_downsample": [
    false,
    true,
    true
  ],
  "z_dim": 16
}