Hugging Face's logo

Efficient-Large-Model
/
Sana_1600M_512px_diffusers

Diffusers
Safetensors
SanaPipeline
Model card Files Community
Sana_1600M_512px_diffusers
File size: 1,280 Bytes 
13a7cc8
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
 
{
  "_class_name": "AutoencoderDC",
  "_diffusers_version": "0.32.0.dev0",
  "_name_or_path": "mit-han-lab/dc-ae-f32c32-sana-1.0-diffusers",
  "attention_head_dim": 32,
  "decoder_act_fns": "silu",
  "decoder_block_out_channels": [
    128,
    256,
    512,
    512,
    1024,
    1024
  ],
  "decoder_block_types": [
    "ResBlock",
    "ResBlock",
    "ResBlock",
    "EfficientViTBlock",
    "EfficientViTBlock",
    "EfficientViTBlock"
  ],
  "decoder_layers_per_block": [
    3,
    3,
    3,
    3,
    3,
    3
  ],
  "decoder_norm_types": "rms_norm",
  "decoder_qkv_multiscales": [
    [],
    [],
    [],
    [
      5
    ],
    [
      5
    ],
    [
      5
    ]
  ],
  "downsample_block_type": "Conv",
  "encoder_block_out_channels": [
    128,
    256,
    512,
    512,
    1024,
    1024
  ],
  "encoder_block_types": [
    "ResBlock",
    "ResBlock",
    "ResBlock",
    "EfficientViTBlock",
    "EfficientViTBlock",
    "EfficientViTBlock"
  ],
  "encoder_layers_per_block": [
    2,
    2,
    2,
    3,
    3,
    3
  ],
  "encoder_qkv_multiscales": [
    [],
    [],
    [],
    [
      5
    ],
    [
      5
    ],
    [
      5
    ]
  ],
  "in_channels": 3,
  "latent_channels": 32,
  "scaling_factor": 0.41407,
  "upsample_block_type": "interpolate"
}