Hugging Face's logo

circulus
/
Wan2.2-I2V-A14B-bnb-diffusers

Diffusers
Safetensors
WanImageToVideoPipeline
Model card Files
xet
Community
rippertnt commited on
Commit
a5ce85d
·
verified ·
1 Parent(s): 4ca29e9

Upload 8 files

Browse files
Files changed (8) hide show
  1. text_encoder/config.json +48 -0
  2. text_encoder/model-00001-of-00002.safetensors +3 -0
  3. text_encoder/model-00002-of-00002.safetensors +3 -0
  4. text_encoder/model.safetensors.index.json +0 -0
  5. transformer/config.json +40 -0
  6. transformer/diffusion_pytorch_model.safetensors +3 -0
  7. transformer_2/config.json +40 -0
  8. transformer_2/diffusion_pytorch_model.safetensors +3 -0
text_encoder/config.json ADDED
@@ -0,0 +1,48 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "UMT5EncoderModel"
4
+ ],
5
+ "classifier_dropout": 0.0,
6
+ "d_ff": 10240,
7
+ "d_kv": 64,
8
+ "d_model": 4096,
9
+ "decoder_start_token_id": 0,
10
+ "dense_act_fn": "gelu_new",
11
+ "dropout_rate": 0.1,
12
+ "dtype": "bfloat16",
13
+ "eos_token_id": 1,
14
+ "feed_forward_proj": "gated-gelu",
15
+ "initializer_factor": 1.0,
16
+ "is_encoder_decoder": true,
17
+ "is_gated_act": true,
18
+ "layer_norm_epsilon": 1e-06,
19
+ "model_type": "umt5",
20
+ "num_decoder_layers": 24,
21
+ "num_heads": 64,
22
+ "num_layers": 24,
23
+ "output_past": true,
24
+ "pad_token_id": 0,
25
+ "quantization_config": {
26
+ "_load_in_4bit": true,
27
+ "_load_in_8bit": false,
28
+ "bnb_4bit_compute_dtype": "bfloat16",
29
+ "bnb_4bit_quant_storage": "uint8",
30
+ "bnb_4bit_quant_type": "nf4",
31
+ "bnb_4bit_use_double_quant": true,
32
+ "llm_int8_enable_fp32_cpu_offload": false,
33
+ "llm_int8_has_fp16_weight": false,
34
+ "llm_int8_skip_modules": null,
35
+ "llm_int8_threshold": 6.0,
36
+ "load_in_4bit": true,
37
+ "load_in_8bit": false,
38
+ "quant_method": "bitsandbytes"
39
+ },
40
+ "relative_attention_max_distance": 128,
41
+ "relative_attention_num_buckets": 32,
42
+ "scalable_attention": true,
43
+ "tie_word_embeddings": false,
44
+ "tokenizer_class": "T5Tokenizer",
45
+ "transformers_version": "4.57.1",
46
+ "use_cache": true,
47
+ "vocab_size": 256384
48
+ }
text_encoder/model-00001-of-00002.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fdb76566b1474dc518efbc46c9f5b6274ca803b4d425be9b7e8b4a65b89b0ed1
3
+ size 4880906532
text_encoder/model-00002-of-00002.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8e4d5ae7464f56d5ac652e5c8176c469db4bef7384decfd27d067f8ecb677d75
3
+ size 3116178803
text_encoder/model.safetensors.index.json ADDED
The diff for this file is too large to render. See raw diff
 
transformer/config.json ADDED
@@ -0,0 +1,40 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_class_name": "WanTransformer3DModel",
3
+ "_diffusers_version": "0.36.0.dev0",
4
+ "_name_or_path": "Wan-AI/Wan2.2-I2V-A14B-Diffusers",
5
+ "added_kv_proj_dim": null,
6
+ "attention_head_dim": 128,
7
+ "cross_attn_norm": true,
8
+ "eps": 1e-06,
9
+ "ffn_dim": 13824,
10
+ "freq_dim": 256,
11
+ "image_dim": null,
12
+ "in_channels": 36,
13
+ "num_attention_heads": 40,
14
+ "num_layers": 40,
15
+ "out_channels": 16,
16
+ "patch_size": [
17
+ 1,
18
+ 2,
19
+ 2
20
+ ],
21
+ "pos_embed_seq_len": null,
22
+ "qk_norm": "rms_norm_across_heads",
23
+ "quantization_config": {
24
+ "_load_in_4bit": true,
25
+ "_load_in_8bit": false,
26
+ "bnb_4bit_compute_dtype": "bfloat16",
27
+ "bnb_4bit_quant_storage": "uint8",
28
+ "bnb_4bit_quant_type": "nf4",
29
+ "bnb_4bit_use_double_quant": true,
30
+ "llm_int8_enable_fp32_cpu_offload": false,
31
+ "llm_int8_has_fp16_weight": false,
32
+ "llm_int8_skip_modules": null,
33
+ "llm_int8_threshold": 6.0,
34
+ "load_in_4bit": true,
35
+ "load_in_8bit": false,
36
+ "quant_method": "bitsandbytes"
37
+ },
38
+ "rope_max_seq_len": 1024,
39
+ "text_dim": 4096
40
+ }
transformer/diffusion_pytorch_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:73fc455049d91ba754a0efb523decf2a8f645bfcb9a4ced454ee764f364e8a3a
3
+ size 7479664332
transformer_2/config.json ADDED
@@ -0,0 +1,40 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_class_name": "WanTransformer3DModel",
3
+ "_diffusers_version": "0.36.0.dev0",
4
+ "_name_or_path": "Wan-AI/Wan2.2-I2V-A14B-Diffusers",
5
+ "added_kv_proj_dim": null,
6
+ "attention_head_dim": 128,
7
+ "cross_attn_norm": true,
8
+ "eps": 1e-06,
9
+ "ffn_dim": 13824,
10
+ "freq_dim": 256,
11
+ "image_dim": null,
12
+ "in_channels": 36,
13
+ "num_attention_heads": 40,
14
+ "num_layers": 40,
15
+ "out_channels": 16,
16
+ "patch_size": [
17
+ 1,
18
+ 2,
19
+ 2
20
+ ],
21
+ "pos_embed_seq_len": null,
22
+ "qk_norm": "rms_norm_across_heads",
23
+ "quantization_config": {
24
+ "_load_in_4bit": true,
25
+ "_load_in_8bit": false,
26
+ "bnb_4bit_compute_dtype": "bfloat16",
27
+ "bnb_4bit_quant_storage": "uint8",
28
+ "bnb_4bit_quant_type": "nf4",
29
+ "bnb_4bit_use_double_quant": true,
30
+ "llm_int8_enable_fp32_cpu_offload": false,
31
+ "llm_int8_has_fp16_weight": false,
32
+ "llm_int8_skip_modules": null,
33
+ "llm_int8_threshold": 6.0,
34
+ "load_in_4bit": true,
35
+ "load_in_8bit": false,
36
+ "quant_method": "bitsandbytes"
37
+ },
38
+ "rope_max_seq_len": 1024,
39
+ "text_dim": 4096
40
+ }
transformer_2/diffusion_pytorch_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:07f22ae091fd0cba3b640a438fb628dbb1a6dfa21be5c72f3615e4ccb6b392b8
3
+ size 7479664368