Hugging Face's logo

Wan-AI
/
Wan2.2-Animate-2-14B-Distilled-Diffusers

Diffusers
Safetensors
WanAnimate2Pipeline
Model card Files
xet
 Community
1

Add diffusers-format transformer weights

#1
by YiYiXu HF Staff - opened
base: refs/heads/main
←
from: refs/pr/1
Discussion Files changed
+1402
-1332
Files changed (9) hide show
  1. model_index.json +1 -1
  2. modular_model_index.json +89 -0
  3. scheduler/scheduler_config.json +10 -22
  4. transformer/config.json +1 -8
  5. transformer/diffusion_pytorch_model-00001-of-00004.safetensors +2 -2
  6. transformer/diffusion_pytorch_model-00002-of-00004.safetensors +2 -2
  7. transformer/diffusion_pytorch_model-00003-of-00004.safetensors +2 -2
  8. transformer/diffusion_pytorch_model-00004-of-00004.safetensors +2 -2
  9. transformer/diffusion_pytorch_model.safetensors.index.json +0 -0
model_index.json CHANGED
@@ -11,7 +11,7 @@
11
  ],
12
  "scheduler": [
13
  "diffusers",
14
- "DPMSolverMultistepScheduler"
15
  ],
16
  "text_encoder": [
17
  "transformers",
 
11
  ],
12
  "scheduler": [
13
  "diffusers",
14
+ "FlowMatchEulerDiscreteScheduler"
15
  ],
16
  "text_encoder": [
17
  "transformers",
modular_model_index.json ADDED
@@ -0,0 +1,89 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_blocks_class_name": "WanAnimate2DistilledBlocks",
3
+ "_class_name": "WanAnimate2DistilledModularPipeline",
4
+ "_diffusers_version": "0.36.0.dev0",
5
+ "image_encoder": [
6
+ "transformers",
7
+ "CLIPVisionModel",
8
+ {
9
+ "pretrained_model_name_or_path": "Wan-AI/Wan2.2-Animate-2-14B-Distilled-Diffusers",
10
+ "revision": null,
11
+ "subfolder": "image_encoder",
12
+ "type_hint": [
13
+ "transformers",
14
+ "CLIPVisionModel"
15
+ ],
16
+ "variant": null
17
+ }
18
+ ],
19
+ "scheduler": [
20
+ "diffusers",
21
+ "FlowMatchEulerDiscreteScheduler",
22
+ {
23
+ "pretrained_model_name_or_path": "Wan-AI/Wan2.2-Animate-2-14B-Distilled-Diffusers",
24
+ "revision": "refs/pr/1",
25
+ "subfolder": "scheduler",
26
+ "type_hint": [
27
+ "diffusers",
28
+ "FlowMatchEulerDiscreteScheduler"
29
+ ],
30
+ "variant": null
31
+ }
32
+ ],
33
+ "text_encoder": [
34
+ "transformers",
35
+ "UMT5EncoderModel",
36
+ {
37
+ "pretrained_model_name_or_path": "Wan-AI/Wan2.2-Animate-2-14B-Distilled-Diffusers",
38
+ "revision": null,
39
+ "subfolder": "text_encoder",
40
+ "type_hint": [
41
+ "transformers",
42
+ "UMT5EncoderModel"
43
+ ],
44
+ "variant": null
45
+ }
46
+ ],
47
+ "tokenizer": [
48
+ "transformers",
49
+ "T5TokenizerFast",
50
+ {
51
+ "pretrained_model_name_or_path": "Wan-AI/Wan2.2-Animate-2-14B-Distilled-Diffusers",
52
+ "revision": null,
53
+ "subfolder": "tokenizer",
54
+ "type_hint": [
55
+ "transformers",
56
+ "T5TokenizerFast"
57
+ ],
58
+ "variant": null
59
+ }
60
+ ],
61
+ "transformer": [
62
+ "diffusers",
63
+ "WanAnimate2Transformer3DModel",
64
+ {
65
+ "pretrained_model_name_or_path": "Wan-AI/Wan2.2-Animate-2-14B-Distilled-Diffusers",
66
+ "revision": "refs/pr/1",
67
+ "subfolder": "transformer",
68
+ "type_hint": [
69
+ "diffusers",
70
+ "WanAnimate2Transformer3DModel"
71
+ ],
72
+ "variant": null
73
+ }
74
+ ],
75
+ "vae": [
76
+ "diffusers",
77
+ "AutoencoderKLWan",
78
+ {
79
+ "pretrained_model_name_or_path": "Wan-AI/Wan2.2-Animate-2-14B-Distilled-Diffusers",
80
+ "revision": null,
81
+ "subfolder": "vae",
82
+ "type_hint": [
83
+ "diffusers",
84
+ "AutoencoderKLWan"
85
+ ],
86
+ "variant": null
87
+ }
88
+ ]
89
+ }
scheduler/scheduler_config.json CHANGED
@@ -1,30 +1,18 @@
1
  {
2
- "_class_name": "UniPCMultistepScheduler",
3
- "_diffusers_version": "0.36.0.dev0",
4
- "beta_end": 0.02,
5
- "beta_schedule": "linear",
6
- "beta_start": 0.0001,
7
- "disable_corrector": [],
8
- "dynamic_thresholding_ratio": 0.995,
9
- "final_sigmas_type": "zero",
10
- "flow_shift": 5.0,
11
- "lower_order_final": true,
12
  "num_train_timesteps": 1000,
13
- "predict_x0": true,
14
- "prediction_type": "flow_prediction",
15
- "rescale_betas_zero_snr": false,
16
- "sample_max_value": 1.0,
17
- "solver_order": 2,
18
- "solver_p": null,
19
- "solver_type": "bh2",
20
- "steps_offset": 0,
21
- "thresholding": false,
22
  "time_shift_type": "exponential",
23
- "timestep_spacing": "linspace",
24
- "trained_betas": null,
25
  "use_beta_sigmas": false,
26
  "use_dynamic_shifting": false,
27
  "use_exponential_sigmas": false,
28
- "use_flow_sigmas": true,
29
  "use_karras_sigmas": false
30
  }
 
1
  {
2
+ "_class_name": "FlowMatchEulerDiscreteScheduler",
3
+ "_diffusers_version": "0.40.0.dev0",
4
+ "base_image_seq_len": 256,
5
+ "base_shift": 0.5,
6
+ "invert_sigmas": false,
7
+ "max_image_seq_len": 4096,
8
+ "max_shift": 1.15,
 
 
 
9
  "num_train_timesteps": 1000,
10
+ "shift": 5.0,
11
+ "shift_terminal": null,
12
+ "stochastic_sampling": false,
 
 
 
 
 
 
13
  "time_shift_type": "exponential",
 
 
14
  "use_beta_sigmas": false,
15
  "use_dynamic_shifting": false,
16
  "use_exponential_sigmas": false,
 
17
  "use_karras_sigmas": false
18
  }
transformer/config.json CHANGED
@@ -7,7 +7,6 @@
7
  "ffn_dim": 13824,
8
  "freq_dim": 256,
9
  "in_dim": 36,
10
- "log_scale": -1.3,
11
  "num_heads": 40,
12
  "num_layers": 40,
13
  "out_dim": 16,
@@ -16,17 +15,11 @@
16
  2,
17
  2
18
  ],
19
- "qk_norm": true,
20
  "refer_offset_h": 0,
21
  "refer_offset_t": 1,
22
  "refer_offset_w": -1,
23
  "refer_stride": 1,
24
- "sparse_type": 0,
25
  "text_dim": 4096,
26
  "text_len": 512,
27
- "use_img_emb": true,
28
- "window_size": [
29
- -1,
30
- -1
31
- ]
32
  }
 
7
  "ffn_dim": 13824,
8
  "freq_dim": 256,
9
  "in_dim": 36,
 
10
  "num_heads": 40,
11
  "num_layers": 40,
12
  "out_dim": 16,
 
15
  2,
16
  2
17
  ],
 
18
  "refer_offset_h": 0,
19
  "refer_offset_t": 1,
20
  "refer_offset_w": -1,
21
  "refer_stride": 1,
 
22
  "text_dim": 4096,
23
  "text_len": 512,
24
+ "use_img_emb": true
 
 
 
 
25
  }
transformer/diffusion_pytorch_model-00001-of-00004.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:83c9f8fe04ae01429c3e46053d1e55e2134fd1942df63ff3fe7b7aa7d2aeaca4
3
- size 9874594352
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ebc182727d948cb61cac929af3b2887e0a4b7c8b683b7356628cf0350d332c0a
3
+ size 9874593104
transformer/diffusion_pytorch_model-00002-of-00004.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:a222285e4eb588d37315c3ccc155a2d33c91c2e93fc7de428005c944ff268f9c
3
- size 9975380624
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e808cf7a7e7aaf358478a2198152f9e3ec8609c66712be397c21b4b7ec7ef501
3
+ size 9975379320
transformer/diffusion_pytorch_model-00003-of-00004.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:0eea9449bfc81f553f804bf2ddd08abeb3ceb043008e3f364b94e1ac5149858c
3
- size 9954402920
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:444eeea631fe6d2df80780f521f6c7716bd079fc177279ba09d2d20479f2dfaf
3
+ size 9954401600
transformer/diffusion_pytorch_model-00004-of-00004.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:eda459bcd5fe68a5a6fea6e758fe0f70466d31c3f53d54fc443aca876884a2d2
3
- size 2985521896
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:130fd0a278a06096e483f08b2492ddd86b89fef8676b3d7b30556b3230a9b005
3
+ size 2985521520
transformer/diffusion_pytorch_model.safetensors.index.json CHANGED
The diff for this file is too large to render. See raw diff