Upload folder using huggingface_hub
Browse files- README.md +7 -3
- model_index.json +2 -2
- text_encoder/config.json +1 -1
- tokenizer/tokenizer_config.json +1 -0
- transformer/config.json +1 -1
- transformer/diffusion_pytorch_model.safetensors +3 -0
- vae/config.json +34 -1
- vae/diffusion_pytorch_model.safetensors +2 -2
README.md
CHANGED
@@ -6,7 +6,7 @@ tags:
|
|
6 |
- image-to-video
|
7 |
---
|
8 |
|
9 |
-
Unofficial Diffusers-format weights for https://huggingface.co/Lightricks/LTX-Video (version 0.9.
|
10 |
|
11 |
Text-to-Video:
|
12 |
|
@@ -15,7 +15,7 @@ import torch
|
|
15 |
from diffusers import LTXPipeline
|
16 |
from diffusers.utils import export_to_video
|
17 |
|
18 |
-
pipe = LTXPipeline.from_pretrained("a-r-r-o-w/LTX-Video-diffusers", torch_dtype=torch.bfloat16)
|
19 |
pipe.to("cuda")
|
20 |
|
21 |
prompt = "A woman with long brown hair and light skin smiles at another woman with long blonde hair. The woman with brown hair wears a black jacket and has a small, barely noticeable mole on her right cheek. The camera angle is a close-up, focused on the woman with brown hair's face. The lighting is warm and natural, likely from the setting sun, casting a soft glow on the scene. The scene appears to be real-life footage"
|
@@ -28,6 +28,8 @@ video = pipe(
|
|
28 |
height=480,
|
29 |
num_frames=161,
|
30 |
num_inference_steps=50,
|
|
|
|
|
31 |
).frames[0]
|
32 |
export_to_video(video, "output.mp4", fps=24)
|
33 |
```
|
@@ -39,7 +41,7 @@ import torch
|
|
39 |
from diffusers import LTXImageToVideoPipeline
|
40 |
from diffusers.utils import export_to_video, load_image
|
41 |
|
42 |
-
pipe = LTXImageToVideoPipeline.from_pretrained("a-r-r-o-w/LTX-Video-diffusers", torch_dtype=torch.bfloat16)
|
43 |
pipe.to("cuda")
|
44 |
|
45 |
image = load_image(
|
@@ -56,6 +58,8 @@ video = pipe(
|
|
56 |
height=480,
|
57 |
num_frames=161,
|
58 |
num_inference_steps=50,
|
|
|
|
|
59 |
).frames[0]
|
60 |
export_to_video(video, "output.mp4", fps=24)
|
61 |
```
|
|
|
6 |
- image-to-video
|
7 |
---
|
8 |
|
9 |
+
Unofficial Diffusers-format weights for https://huggingface.co/Lightricks/LTX-Video (version 0.9.1).
|
10 |
|
11 |
Text-to-Video:
|
12 |
|
|
|
15 |
from diffusers import LTXPipeline
|
16 |
from diffusers.utils import export_to_video
|
17 |
|
18 |
+
pipe = LTXPipeline.from_pretrained("a-r-r-o-w/LTX-Video-0.9.1-diffusers", torch_dtype=torch.bfloat16)
|
19 |
pipe.to("cuda")
|
20 |
|
21 |
prompt = "A woman with long brown hair and light skin smiles at another woman with long blonde hair. The woman with brown hair wears a black jacket and has a small, barely noticeable mole on her right cheek. The camera angle is a close-up, focused on the woman with brown hair's face. The lighting is warm and natural, likely from the setting sun, casting a soft glow on the scene. The scene appears to be real-life footage"
|
|
|
28 |
height=480,
|
29 |
num_frames=161,
|
30 |
num_inference_steps=50,
|
31 |
+
decode_timestep=0.03,
|
32 |
+
decode_noise_scale=0.025,
|
33 |
).frames[0]
|
34 |
export_to_video(video, "output.mp4", fps=24)
|
35 |
```
|
|
|
41 |
from diffusers import LTXImageToVideoPipeline
|
42 |
from diffusers.utils import export_to_video, load_image
|
43 |
|
44 |
+
pipe = LTXImageToVideoPipeline.from_pretrained("a-r-r-o-w/LTX-Video-0.9.1-diffusers", torch_dtype=torch.bfloat16)
|
45 |
pipe.to("cuda")
|
46 |
|
47 |
image = load_image(
|
|
|
58 |
height=480,
|
59 |
num_frames=161,
|
60 |
num_inference_steps=50,
|
61 |
+
decode_timestep=0.03,
|
62 |
+
decode_noise_scale=0.025,
|
63 |
).frames[0]
|
64 |
export_to_video(video, "output.mp4", fps=24)
|
65 |
```
|
model_index.json
CHANGED
@@ -15,10 +15,10 @@
|
|
15 |
],
|
16 |
"transformer": [
|
17 |
"diffusers",
|
18 |
-
"
|
19 |
],
|
20 |
"vae": [
|
21 |
"diffusers",
|
22 |
-
"
|
23 |
]
|
24 |
}
|
|
|
15 |
],
|
16 |
"transformer": [
|
17 |
"diffusers",
|
18 |
+
"LTXVideoTransformer3DModel"
|
19 |
],
|
20 |
"vae": [
|
21 |
"diffusers",
|
22 |
+
"AutoencoderKLLTXVideo"
|
23 |
]
|
24 |
}
|
text_encoder/config.json
CHANGED
@@ -26,7 +26,7 @@
|
|
26 |
"relative_attention_num_buckets": 32,
|
27 |
"tie_word_embeddings": false,
|
28 |
"torch_dtype": "float32",
|
29 |
-
"transformers_version": "4.
|
30 |
"use_cache": true,
|
31 |
"vocab_size": 32128
|
32 |
}
|
|
|
26 |
"relative_attention_num_buckets": 32,
|
27 |
"tie_word_embeddings": false,
|
28 |
"torch_dtype": "float32",
|
29 |
+
"transformers_version": "4.48.0.dev0",
|
30 |
"use_cache": true,
|
31 |
"vocab_size": 32128
|
32 |
}
|
tokenizer/tokenizer_config.json
CHANGED
@@ -931,6 +931,7 @@
|
|
931 |
"clean_up_tokenization_spaces": false,
|
932 |
"eos_token": "</s>",
|
933 |
"extra_ids": 100,
|
|
|
934 |
"legacy": true,
|
935 |
"model_max_length": 128,
|
936 |
"pad_token": "<pad>",
|
|
|
931 |
"clean_up_tokenization_spaces": false,
|
932 |
"eos_token": "</s>",
|
933 |
"extra_ids": 100,
|
934 |
+
"extra_special_tokens": {},
|
935 |
"legacy": true,
|
936 |
"model_max_length": 128,
|
937 |
"pad_token": "<pad>",
|
transformer/config.json
CHANGED
@@ -1,5 +1,5 @@
|
|
1 |
{
|
2 |
-
"_class_name": "
|
3 |
"_diffusers_version": "0.32.0.dev0",
|
4 |
"activation_fn": "gelu-approximate",
|
5 |
"attention_bias": true,
|
|
|
1 |
{
|
2 |
+
"_class_name": "LTXVideoTransformer3DModel",
|
3 |
"_diffusers_version": "0.32.0.dev0",
|
4 |
"activation_fn": "gelu-approximate",
|
5 |
"attention_bias": true,
|
transformer/diffusion_pytorch_model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:75e214999dbee61d5b8dd3d865dd4927c501f71510ab4efe612db1b50f96b973
|
3 |
+
size 3846852608
|
vae/config.json
CHANGED
@@ -1,5 +1,5 @@
|
|
1 |
{
|
2 |
-
"_class_name": "
|
3 |
"_diffusers_version": "0.32.0.dev0",
|
4 |
"block_out_channels": [
|
5 |
128,
|
@@ -7,7 +7,29 @@
|
|
7 |
512,
|
8 |
512
|
9 |
],
|
|
|
|
|
|
|
|
|
|
|
10 |
"decoder_causal": false,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
11 |
"encoder_causal": true,
|
12 |
"in_channels": 3,
|
13 |
"latent_channels": 128,
|
@@ -28,5 +50,16 @@
|
|
28 |
true,
|
29 |
true,
|
30 |
false
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
31 |
]
|
32 |
}
|
|
|
1 |
{
|
2 |
+
"_class_name": "AutoencoderKLLTXVideo",
|
3 |
"_diffusers_version": "0.32.0.dev0",
|
4 |
"block_out_channels": [
|
5 |
128,
|
|
|
7 |
512,
|
8 |
512
|
9 |
],
|
10 |
+
"decoder_block_out_channels": [
|
11 |
+
256,
|
12 |
+
512,
|
13 |
+
1024
|
14 |
+
],
|
15 |
"decoder_causal": false,
|
16 |
+
"decoder_inject_noise": [
|
17 |
+
true,
|
18 |
+
true,
|
19 |
+
true,
|
20 |
+
false
|
21 |
+
],
|
22 |
+
"decoder_layers_per_block": [
|
23 |
+
5,
|
24 |
+
6,
|
25 |
+
7,
|
26 |
+
8
|
27 |
+
],
|
28 |
+
"decoder_spatio_temporal_scaling": [
|
29 |
+
true,
|
30 |
+
true,
|
31 |
+
true
|
32 |
+
],
|
33 |
"encoder_causal": true,
|
34 |
"in_channels": 3,
|
35 |
"latent_channels": 128,
|
|
|
50 |
true,
|
51 |
true,
|
52 |
false
|
53 |
+
],
|
54 |
+
"timestep_conditioning": true,
|
55 |
+
"upsample_factor": [
|
56 |
+
2,
|
57 |
+
2,
|
58 |
+
2
|
59 |
+
],
|
60 |
+
"upsample_residual": [
|
61 |
+
true,
|
62 |
+
true,
|
63 |
+
true
|
64 |
]
|
65 |
}
|
vae/diffusion_pytorch_model.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:3a70d8d49fb2cc3698ffe9ed0e09fba5cd65d6a7d83fe89320ebc1d6fcc94536
|
3 |
+
size 1869989690
|