initial upload: convert x1.5 from Lightricks/LTX-2.3
Browse files- README.md +35 -0
- latent_upsampler/config.json +12 -0
- latent_upsampler/diffusion_pytorch_model.safetensors +3 -0
- model_index.json +8 -0
README.md
ADDED
|
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
library_name: diffusers
|
| 4 |
+
tags:
|
| 5 |
+
- video
|
| 6 |
+
- ltx-2
|
| 7 |
+
- ltx-2.3
|
| 8 |
+
- latent-upsampler
|
| 9 |
+
base_model: Lightricks/LTX-2.3
|
| 10 |
+
---
|
| 11 |
+
|
| 12 |
+
# LTX-2.3 Spatial Upscaler x1.5 1.0 (Diffusers)
|
| 13 |
+
|
| 14 |
+
Diffusers-format conversion of [`ltx-2.3-spatial-upscaler-x1.5-1.0.safetensors`](https://huggingface.co/Lightricks/LTX-2.3/blob/main/ltx-2.3-spatial-upscaler-x1.5-1.0.safetensors) from the official [Lightricks/LTX-2.3](https://huggingface.co/Lightricks/LTX-2.3) release.
|
| 15 |
+
|
| 16 |
+
Weights are byte-identical to the upstream release; only the layout has been rearranged so `LTX2LatentUpsamplerModel.from_pretrained` and `LTX2LatentUpsamplePipeline` can consume it directly.
|
| 17 |
+
|
| 18 |
+
## Usage
|
| 19 |
+
|
| 20 |
+
```python
|
| 21 |
+
import torch
|
| 22 |
+
from diffusers.pipelines.ltx2 import LTX2LatentUpsamplerModel
|
| 23 |
+
from diffusers.pipelines.ltx2.pipeline_ltx2_latent_upsample import LTX2LatentUpsamplePipeline
|
| 24 |
+
|
| 25 |
+
latent_upsampler = LTX2LatentUpsamplerModel.from_pretrained(
|
| 26 |
+
'<your-username>/<your-repo>',
|
| 27 |
+
subfolder='latent_upsampler',
|
| 28 |
+
torch_dtype=torch.bfloat16,
|
| 29 |
+
)
|
| 30 |
+
upsample_pipe = LTX2LatentUpsamplePipeline(vae=your_pipe.vae, latent_upsampler=latent_upsampler)
|
| 31 |
+
```
|
| 32 |
+
|
| 33 |
+
## License
|
| 34 |
+
|
| 35 |
+
Apache 2.0, inherited from the upstream Lightricks/LTX-2.3 release.
|
latent_upsampler/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_class_name": "LTX2LatentUpsamplerModel",
|
| 3 |
+
"_diffusers_version": "0.38.0.dev0",
|
| 4 |
+
"dims": 3,
|
| 5 |
+
"in_channels": 128,
|
| 6 |
+
"mid_channels": 1024,
|
| 7 |
+
"num_blocks_per_stage": 4,
|
| 8 |
+
"rational_spatial_scale": 1.5,
|
| 9 |
+
"spatial_upsample": true,
|
| 10 |
+
"temporal_upsample": false,
|
| 11 |
+
"use_rational_resampler": true
|
| 12 |
+
}
|
latent_upsampler/diffusion_pytorch_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8d376f6e493b46cc0496913a0b9a55faa004336b24bc8e0ec6de653c9a1d3f9b
|
| 3 |
+
size 1090125562
|
model_index.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_class_name": "LTX2LatentUpsamplePipeline",
|
| 3 |
+
"_diffusers_version": "0.37.0.dev0",
|
| 4 |
+
"latent_upsampler": [
|
| 5 |
+
"ltx2",
|
| 6 |
+
"LTX2LatentUpsamplerModel"
|
| 7 |
+
]
|
| 8 |
+
}
|