Hugging Face's logo

TensorStack
/
Components

Diffusers
Safetensors
Model card Files
xet
 Community
saddam213 commited on
Commit
d555214
·
verified ·
1 Parent(s): 44289bd

Upload 6 files

Browse files
Files changed (6) hide show
  1. AceStepXL_AudioDetokenizer/config.json +21 -0
  2. AceStepXL_AudioDetokenizer/diffusion_pytorch_model.safetensors +3 -0
  3. AceStepXL_AudioTokenizer/config.json +31 -0
  4. AceStepXL_AudioTokenizer/diffusion_pytorch_model.safetensors +3 -0
  5. AceStepXL_ConditionEncoder/config.json +19 -0
  6. AceStepXL_ConditionEncoder/diffusion_pytorch_model.safetensors +3 -0
AceStepXL_AudioDetokenizer/config.json ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_class_name": "AceStepAudioTokenDetokenizer",
3
+ "_diffusers_version": "0.38.0.dev0",
4
+ "attention_bias": false,
5
+ "attention_dropout": 0.0,
6
+ "audio_acoustic_hidden_dim": 64,
7
+ "head_dim": 128,
8
+ "hidden_size": 2048,
9
+ "intermediate_size": 6144,
10
+ "layer_types": [
11
+ "sliding_attention",
12
+ "full_attention"
13
+ ],
14
+ "num_attention_heads": 16,
15
+ "num_attention_pooler_hidden_layers": 2,
16
+ "num_key_value_heads": 8,
17
+ "pool_window_size": 5,
18
+ "rms_norm_eps": 1e-06,
19
+ "rope_theta": 1000000,
20
+ "sliding_window": 128
21
+ }
AceStepXL_AudioDetokenizer/diffusion_pytorch_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:744612a9406114fc5b3661e3682f9bd835dc737538b9777124c874303bbbb837
3
+ size 210026424
AceStepXL_AudioTokenizer/config.json ADDED
@@ -0,0 +1,31 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_class_name": "AceStepAudioTokenizer",
3
+ "_diffusers_version": "0.38.0.dev0",
4
+ "attention_bias": false,
5
+ "attention_dropout": 0.0,
6
+ "audio_acoustic_hidden_dim": 64,
7
+ "fsq_dim": 2048,
8
+ "fsq_input_levels": [
9
+ 8,
10
+ 8,
11
+ 8,
12
+ 5,
13
+ 5,
14
+ 5
15
+ ],
16
+ "fsq_input_num_quantizers": 1,
17
+ "head_dim": 128,
18
+ "hidden_size": 2048,
19
+ "intermediate_size": 6144,
20
+ "layer_types": [
21
+ "sliding_attention",
22
+ "full_attention"
23
+ ],
24
+ "num_attention_heads": 16,
25
+ "num_attention_pooler_hidden_layers": 2,
26
+ "num_key_value_heads": 8,
27
+ "pool_window_size": 5,
28
+ "rms_norm_eps": 1e-06,
29
+ "rope_theta": 1000000,
30
+ "sliding_window": 128
31
+ }
AceStepXL_AudioTokenizer/diffusion_pytorch_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f8f46814109ba8bb451fef2122cb7ff58093856a9f2e5464e32f06b71404973b
3
+ size 210068124
AceStepXL_ConditionEncoder/config.json ADDED
@@ -0,0 +1,19 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_class_name": "AceStepConditionEncoder",
3
+ "_diffusers_version": "0.38.0.dev0",
4
+ "attention_bias": false,
5
+ "attention_dropout": 0.0,
6
+ "head_dim": 128,
7
+ "hidden_size": 2048,
8
+ "intermediate_size": 6144,
9
+ "layer_types": null,
10
+ "num_attention_heads": 16,
11
+ "num_key_value_heads": 8,
12
+ "num_lyric_encoder_hidden_layers": 8,
13
+ "num_timbre_encoder_hidden_layers": 4,
14
+ "rms_norm_eps": 1e-06,
15
+ "rope_theta": 1000000,
16
+ "sliding_window": 128,
17
+ "text_hidden_dim": 1024,
18
+ "timbre_hidden_dim": 64
19
+ }
AceStepXL_ConditionEncoder/diffusion_pytorch_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ca111ca95858c8d3c3678c986dda03df20778c0de009f57109c1052866d522b2
3
+ size 1218676312