Hugging Face's logo

optimum-intel-internal-testing
/
tiny-random-ltx2.5

Diffusers
Safetensors
LTX2Pipeline
Model card Files
xet
 Community
1
h0witended commited on
Commit
5e56afb
·
verified ·
1 Parent(s): 9e88871

Rebuild fixture with diffusers 0.40.0 (ff_bias, use_keyframes_abs_pos_embedding, use_prompt_adaln_single) for architecture fidelity with real LTX-2.5

Browse files
Files changed (12) hide show
  1. audio_vae/config.json +1 -1
  2. audio_vae/diffusion_pytorch_model.safetensors +1 -1
  3. connectors/config.json +1 -1
  4. connectors/diffusion_pytorch_model.safetensors +1 -1
  5. model_index.json +9 -1
  6. scheduler/scheduler_config.json +1 -1
  7. transformer/config.json +5 -1
  8. transformer/diffusion_pytorch_model.safetensors +2 -2
  9. vae/config.json +1 -1
  10. vae/diffusion_pytorch_model.safetensors +1 -1
  11. vocoder/config.json +1 -1
  12. vocoder/diffusion_pytorch_model.safetensors +1 -1
audio_vae/config.json CHANGED
@@ -1,6 +1,6 @@
1
  {
2
  "_class_name": "AutoencoderKLLTX2Audio",
3
- "_diffusers_version": "0.39.0",
4
  "attn_resolutions": null,
5
  "base_channels": 128,
6
  "causality_axis": "height",
 
1
  {
2
  "_class_name": "AutoencoderKLLTX2Audio",
3
+ "_diffusers_version": "0.40.0",
4
  "attn_resolutions": null,
5
  "base_channels": 128,
6
  "causality_axis": "height",
audio_vae/diffusion_pytorch_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:caa96021f24c030ff2174d17fab163fe69ce886abff98946b844e8eee15db494
3
  size 8399760
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4879d6b1157d97f25ba9c8dc8b74258892f7fd3c43447e6ab329e4db571a237f
3
  size 8399760
connectors/config.json CHANGED
@@ -1,6 +1,6 @@
1
  {
2
  "_class_name": "LTX2TextConnectors",
3
- "_diffusers_version": "0.39.0",
4
  "audio_connector_attention_head_dim": 8,
5
  "audio_connector_num_attention_heads": 4,
6
  "audio_connector_num_layers": 1,
 
1
  {
2
  "_class_name": "LTX2TextConnectors",
3
+ "_diffusers_version": "0.40.0",
4
  "audio_connector_attention_head_dim": 8,
5
  "audio_connector_num_attention_heads": 4,
6
  "audio_connector_num_layers": 1,
connectors/diffusion_pytorch_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:fee2082b6d1307ae82d41624a9e143ba93bea20dd80157c67e67c32772527228
3
  size 133392
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:195116083dd54d3ef7f83fa39fab401005de53b76729c83393a2d00d2a7929fd
3
  size 133392
model_index.json CHANGED
@@ -1,6 +1,6 @@
1
  {
2
  "_class_name": "LTX2Pipeline",
3
- "_diffusers_version": "0.39.0",
4
  "audio_vae": [
5
  "diffusers",
6
  "AutoencoderKLLTX2Audio"
@@ -9,10 +9,18 @@
9
  "ltx2",
10
  "LTX2TextConnectors"
11
  ],
 
 
 
 
12
  "processor": [
13
  null,
14
  null
15
  ],
 
 
 
 
16
  "scheduler": [
17
  "diffusers",
18
  "FlowMatchEulerDiscreteScheduler"
 
1
  {
2
  "_class_name": "LTX2Pipeline",
3
+ "_diffusers_version": "0.40.0",
4
  "audio_vae": [
5
  "diffusers",
6
  "AutoencoderKLLTX2Audio"
 
9
  "ltx2",
10
  "LTX2TextConnectors"
11
  ],
12
+ "duration_head": [
13
+ null,
14
+ null
15
+ ],
16
  "processor": [
17
  null,
18
  null
19
  ],
20
+ "prompt_enhancer": [
21
+ null,
22
+ null
23
+ ],
24
  "scheduler": [
25
  "diffusers",
26
  "FlowMatchEulerDiscreteScheduler"
scheduler/scheduler_config.json CHANGED
@@ -1,6 +1,6 @@
1
  {
2
  "_class_name": "FlowMatchEulerDiscreteScheduler",
3
- "_diffusers_version": "0.39.0",
4
  "base_image_seq_len": 1024,
5
  "base_shift": 0.95,
6
  "invert_sigmas": false,
 
1
  {
2
  "_class_name": "FlowMatchEulerDiscreteScheduler",
3
+ "_diffusers_version": "0.40.0",
4
  "base_image_seq_len": 1024,
5
  "base_shift": 0.95,
6
  "invert_sigmas": false,
transformer/config.json CHANGED
@@ -1,6 +1,6 @@
1
  {
2
  "_class_name": "LTX2VideoTransformer3DModel",
3
- "_diffusers_version": "0.39.0",
4
  "activation_fn": "gelu-approximate",
5
  "attention_bias": true,
6
  "attention_head_dim": 8,
@@ -8,6 +8,7 @@
8
  "audio_attention_head_dim": 8,
9
  "audio_cross_attention_dim": 32,
10
  "audio_cross_attn_mod": true,
 
11
  "audio_gated_attn": true,
12
  "audio_hop_length": 160,
13
  "audio_in_channels": 128,
@@ -25,6 +26,7 @@
25
  "cross_attention_dim": 32,
26
  "cross_attn_mod": true,
27
  "cross_attn_timestep_scale_multiplier": 1000,
 
28
  "gated_attn": true,
29
  "in_channels": 8,
30
  "norm_elementwise_affine": false,
@@ -41,6 +43,8 @@
41
  "rope_theta": 10000.0,
42
  "rope_type": "split",
43
  "timestep_scale_multiplier": 1000,
 
 
44
  "use_prompt_embeddings": false,
45
  "vae_scale_factors": [
46
  8,
 
1
  {
2
  "_class_name": "LTX2VideoTransformer3DModel",
3
+ "_diffusers_version": "0.40.0",
4
  "activation_fn": "gelu-approximate",
5
  "attention_bias": true,
6
  "attention_head_dim": 8,
 
8
  "audio_attention_head_dim": 8,
9
  "audio_cross_attention_dim": 32,
10
  "audio_cross_attn_mod": true,
11
+ "audio_ff_bias": true,
12
  "audio_gated_attn": true,
13
  "audio_hop_length": 160,
14
  "audio_in_channels": 128,
 
26
  "cross_attention_dim": 32,
27
  "cross_attn_mod": true,
28
  "cross_attn_timestep_scale_multiplier": 1000,
29
+ "ff_bias": false,
30
  "gated_attn": true,
31
  "in_channels": 8,
32
  "norm_elementwise_affine": false,
 
43
  "rope_theta": 10000.0,
44
  "rope_type": "split",
45
  "timestep_scale_multiplier": 1000,
46
+ "use_keyframes_abs_pos_embedding": true,
47
+ "use_prompt_adaln_single": true,
48
  "use_prompt_embeddings": false,
49
  "vae_scale_factors": [
50
  8,
transformer/diffusion_pytorch_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:2f6826c55a4b348fc0f7258b958f793bd1fde4e5146b644cf109b15e882cb264
3
- size 847608
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4a84d5174365db1cb1697413aacb2428231bc0c3c1387972991a63c5fd42e195
3
+ size 846144
vae/config.json CHANGED
@@ -1,6 +1,6 @@
1
  {
2
  "_class_name": "AutoencoderKLLTX2Video",
3
- "_diffusers_version": "0.39.0",
4
  "audio_latents_mean_data": [
5
  0.0,
6
  0.0,
 
1
  {
2
  "_class_name": "AutoencoderKLLTX2Video",
3
+ "_diffusers_version": "0.40.0",
4
  "audio_latents_mean_data": [
5
  0.0,
6
  0.0,
vae/diffusion_pytorch_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:8e29e9a5e84763284f7142334d7f32730c1ea34faaf3e89d85c2317d09068248
3
  size 1056860
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:23aa7e896e48f467bd088b00d4a95cee1efba97483b5b8ace3210e4a23f7fe93
3
  size 1056860
vocoder/config.json CHANGED
@@ -1,6 +1,6 @@
1
  {
2
  "_class_name": "LTX2VocoderWithBWE",
3
- "_diffusers_version": "0.39.0",
4
  "act_fn": "snakebeta",
5
  "antialias": true,
6
  "antialias_kernel_size": 12,
 
1
  {
2
  "_class_name": "LTX2VocoderWithBWE",
3
+ "_diffusers_version": "0.40.0",
4
  "act_fn": "snakebeta",
5
  "antialias": true,
6
  "antialias_kernel_size": 12,
vocoder/diffusion_pytorch_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f2a333fa30478e22d49f57fc689486fa0a6b76e3c28376cccf0a53f247d7a4fe
3
  size 2829208
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:89b083230b9eaa8ce3dc583ce5806de3f86fa253ca5c1f33d17bc9942eb04bc7
3
  size 2829208