Hugging Face's logo

windecay
/
SimpleSDXL2

ONNX
Safetensors
Model card Files
xet
 Community
windecay commited on
Commit
3dbb380
·
verified ·
1 Parent(s): fc2d4f0

Upload 10 files

Browse files
Files changed (10) hide show
  1. SimpleModels/hunyuan_foley/clap/config.json +29 -0
  2. SimpleModels/hunyuan_foley/clap/merges.txt +0 -0
  3. SimpleModels/hunyuan_foley/clap/pytorch_model.bin +3 -0
  4. SimpleModels/hunyuan_foley/clap/vocab.json +0 -0
  5. SimpleModels/hunyuan_foley/hunyuanvideo_foley_xl.pth +3 -0
  6. SimpleModels/hunyuan_foley/siglip2/config.json +13 -0
  7. SimpleModels/hunyuan_foley/siglip2/model.safetensors +3 -0
  8. SimpleModels/hunyuan_foley/siglip2/preprocessor_config.json +24 -0
  9. SimpleModels/hunyuan_foley/synchformer_state_dict.pth +3 -0
  10. SimpleModels/hunyuan_foley/vae_128d_48k.pth +3 -0
SimpleModels/hunyuan_foley/clap/config.json ADDED
@@ -0,0 +1,29 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "/home/yoach/tmp/clap/larger_clap_music_speech_general",
3
+ "architectures": [
4
+ "ClapModel"
5
+ ],
6
+ "audio_config": {
7
+ "depths": [
8
+ 2,
9
+ 2,
10
+ 12,
11
+ 2
12
+ ],
13
+ "hidden_size": 1024,
14
+ "model_type": "clap_audio_model",
15
+ "patch_embeds_hidden_size": 128
16
+ },
17
+ "hidden_size": 768,
18
+ "initializer_factor": 1.0,
19
+ "logit_scale_init_value": 14.285714285714285,
20
+ "model_type": "clap",
21
+ "num_hidden_layers": 16,
22
+ "projection_dim": 512,
23
+ "projection_hidden_act": "relu",
24
+ "text_config": {
25
+ "model_type": "clap_text_model"
26
+ },
27
+ "torch_dtype": "float32",
28
+ "transformers_version": "4.35.0.dev0"
29
+ }
SimpleModels/hunyuan_foley/clap/merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
SimpleModels/hunyuan_foley/clap/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:314eb00cce6ad68d25237b8446b659ccdb136ed4672c1bca470f142f72455026
3
+ size 776444665
SimpleModels/hunyuan_foley/clap/vocab.json ADDED
The diff for this file is too large to render. See raw diff
 
SimpleModels/hunyuan_foley/hunyuanvideo_foley_xl.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cf7e93dca93063942bc22b53d4fe888e9d9d5a1ad7977873b7b2c6b126cf00ca
3
+ size 5854140970
SimpleModels/hunyuan_foley/siglip2/config.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "initializer_factor": 1.0,
3
+ "model_type": "siglip",
4
+ "text_config": {
5
+ "model_type": "siglip_text_model",
6
+ "vocab_size": 256000
7
+ },
8
+ "transformers_version": "4.49.0.dev0",
9
+ "vision_config": {
10
+ "image_size": 512,
11
+ "model_type": "siglip_vision_model"
12
+ }
13
+ }
SimpleModels/hunyuan_foley/siglip2/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fe0e601c625e69eed8e73500d39e9b6164403fe03db8048e87913c3cefbbb3fe
3
+ size 1503344520
SimpleModels/hunyuan_foley/siglip2/preprocessor_config.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_convert_rgb": null,
3
+ "do_normalize": true,
4
+ "do_rescale": true,
5
+ "do_resize": true,
6
+ "image_mean": [
7
+ 0.5,
8
+ 0.5,
9
+ 0.5
10
+ ],
11
+ "image_processor_type": "SiglipImageProcessor",
12
+ "image_std": [
13
+ 0.5,
14
+ 0.5,
15
+ 0.5
16
+ ],
17
+ "processor_class": "SiglipProcessor",
18
+ "resample": 2,
19
+ "rescale_factor": 0.00392156862745098,
20
+ "size": {
21
+ "height": 512,
22
+ "width": 512
23
+ }
24
+ }
SimpleModels/hunyuan_foley/synchformer_state_dict.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8aff082f2df5c3bc52759db0c865c7ee772ae6400b860d1b7e90413f2defb67c
3
+ size 950058171
SimpleModels/hunyuan_foley/vae_128d_48k.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:07e6139ff33bd21ba8a7a7f40ed24aab13ac0d100cd686f8a9f3e03dc5251cb1
3
+ size 1486465965