Instructions to use PrunaAI/PrunaVAED with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use PrunaAI/PrunaVAED with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("PrunaAI/PrunaVAED", torch_dtype=torch.bfloat16, device_map="cuda") prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k" image = pipe(prompt).images[0] - LTX.io
How to use PrunaAI/PrunaVAED with LTX.io:
# Install the LTX-2 pipelines git clone https://github.com/Lightricks/LTX-2.git cd LTX-2 uv sync --frozen
# Download the weights from this repo, plus the Gemma text encoder hf download PrunaAI/PrunaVAED --local-dir models/PrunaVAED hf download google/gemma-3-12b-it-qat-q4_0-unquantized --local-dir models/gemma-3-12b
# Fast pipeline (distilled model, no distilled LoRA needed) uv run python -m ltx_pipelines.distilled \ --distilled-checkpoint-path models/PrunaVAED/<distilled-checkpoint>.safetensors \ --spatial-upsampler-path models/PrunaVAED/<spatial-upsampler>.safetensors \ --gemma-root models/gemma-3-12b \ --prompt "A beautiful sunset over the ocean" \ --output-path output.mp4 # For image-to-video, add: --image path/to/image.jpg 0 0.8# HQ pipeline (two-stage, higher quality) uv run python -m ltx_pipelines.ti2vid_two_stages_hq \ --checkpoint-path models/PrunaVAED/<checkpoint>.safetensors \ --distilled-lora models/PrunaVAED/<distilled-lora>.safetensors 0.8 \ --spatial-upsampler-path models/PrunaVAED/<spatial-upsampler>.safetensors \ --gemma-root models/gemma-3-12b \ --prompt "A beautiful sunset over the ocean" \ --output-path output.mp4 # For image-to-video, add: --image path/to/image.jpg 0 0.8 - Notebooks
- Google Colab
- Kaggle
Update README.md
Browse files- Updated an issue with tables. They weren't correctly rendered, so I just reset to the previous visualization.
- Updated the CTA
README.md
CHANGED
|
@@ -127,7 +127,7 @@ pipeline and the
|
|
| 127 |
[`distilled`](https://github.com/Lightricks/LTX-2/blob/main/packages/ltx-pipelines/src/ltx_pipelines/distilled.py)
|
| 128 |
pipeline (recommended). (~4 s @ 24 fps)
|
| 129 |
|
| 130 |
-
<table
|
| 131 |
<tr><th>Setting</th><th>Value</th></tr>
|
| 132 |
<tr><td>Model</td><td>PrunaVAED</td></tr>
|
| 133 |
<tr><td>Precision</td><td><code>bfloat16</code></td></tr>
|
|
@@ -136,7 +136,7 @@ pipeline (recommended). (~4 s @ 24 fps)
|
|
| 136 |
<tr><td>Decode</td><td>Full (no tiling / rolling / <code>torch.compile</code>)</td></tr>
|
| 137 |
</table>
|
| 138 |
|
| 139 |
-
<table
|
| 140 |
<tr>
|
| 141 |
<th>Clip length</th><th>LTX-2.3 VAED</th><th>PrunaVAED</th><th>Speedup</th>
|
| 142 |
</tr>
|
|
@@ -156,7 +156,7 @@ pipeline (recommended). (~4 s @ 24 fps)
|
|
| 156 |
|
| 157 |
\* Tiling required on H100; 14533ms otherwise
|
| 158 |
|
| 159 |
-
<table
|
| 160 |
<tr>
|
| 161 |
<th>Latent set</th><th>Clips</th><th>PSNR ↑</th><th>LPIPS ↓</th><th>SSIM ↑</th><th>Δ-frame PSNR ↑</th>
|
| 162 |
</tr>
|
|
@@ -247,9 +247,40 @@ and commercial terms before using or redistributing the model.
|
|
| 247 |
Helper code adapted from Hugging Face diffusers retains its Apache-2.0
|
| 248 |
attribution; see [`NOTICE`](NOTICE).
|
| 249 |
|
| 250 |
-
##
|
| 251 |
|
| 252 |
- **Use PrunaVAED to speed up LTX-2.3 decoding.**
|
| 253 |
-
-
|
|
|
|
| 254 |
[P-Video](https://www.pruna.ai/p-video) and
|
| 255 |
-
[P-Video documentation](https://docs.pruna.ai/en/stable/docs_pruna_endpoints/performance_models/p-video.html).
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 127 |
[`distilled`](https://github.com/Lightricks/LTX-2/blob/main/packages/ltx-pipelines/src/ltx_pipelines/distilled.py)
|
| 128 |
pipeline (recommended). (~4 s @ 24 fps)
|
| 129 |
|
| 130 |
+
<table class="w-full table-fixed">
|
| 131 |
<tr><th>Setting</th><th>Value</th></tr>
|
| 132 |
<tr><td>Model</td><td>PrunaVAED</td></tr>
|
| 133 |
<tr><td>Precision</td><td><code>bfloat16</code></td></tr>
|
|
|
|
| 136 |
<tr><td>Decode</td><td>Full (no tiling / rolling / <code>torch.compile</code>)</td></tr>
|
| 137 |
</table>
|
| 138 |
|
| 139 |
+
<table class="w-full table-fixed">
|
| 140 |
<tr>
|
| 141 |
<th>Clip length</th><th>LTX-2.3 VAED</th><th>PrunaVAED</th><th>Speedup</th>
|
| 142 |
</tr>
|
|
|
|
| 156 |
|
| 157 |
\* Tiling required on H100; 14533ms otherwise
|
| 158 |
|
| 159 |
+
<table class="w-full table-fixed">
|
| 160 |
<tr>
|
| 161 |
<th>Latent set</th><th>Clips</th><th>PSNR ↑</th><th>LPIPS ↓</th><th>SSIM ↑</th><th>Δ-frame PSNR ↑</th>
|
| 162 |
</tr>
|
|
|
|
| 247 |
Helper code adapted from Hugging Face diffusers retains its Apache-2.0
|
| 248 |
attribution; see [`NOTICE`](NOTICE).
|
| 249 |
|
| 250 |
+
## What's next?
|
| 251 |
|
| 252 |
- **Use PrunaVAED to speed up LTX-2.3 decoding.**
|
| 253 |
+
- Compress your own models with [Pruna](https://github.com/PrunaAI/pruna) and give us a ⭐️ for more efficiency!
|
| 254 |
+
- Want to use our optimized video model right away? Check
|
| 255 |
[P-Video](https://www.pruna.ai/p-video) and
|
| 256 |
+
[P-Video documentation](https://docs.pruna.ai/en/stable/docs_pruna_endpoints/performance_models/p-video.html).
|
| 257 |
+
|
| 258 |
+
<style>
|
| 259 |
+
.model-button {
|
| 260 |
+
display: inline-flex;
|
| 261 |
+
flex-direction: row;
|
| 262 |
+
justify-content: center;
|
| 263 |
+
align-items: center;
|
| 264 |
+
gap: 8px;
|
| 265 |
+
|
| 266 |
+
padding: 8px 20px;
|
| 267 |
+
border: none;
|
| 268 |
+
border-radius: 8px;
|
| 269 |
+
|
| 270 |
+
background: #9334e9;
|
| 271 |
+
color: #ffffff;
|
| 272 |
+
|
| 273 |
+
font-size: 14px;
|
| 274 |
+
font-weight: 400;
|
| 275 |
+
line-height: 1;
|
| 276 |
+
text-decoration: none;
|
| 277 |
+
white-space: nowrap;
|
| 278 |
+
cursor: pointer;
|
| 279 |
+
|
| 280 |
+
box-sizing: border-box;
|
| 281 |
+
overflow: visible;
|
| 282 |
+
opacity: 1;
|
| 283 |
+
}
|
| 284 |
+
</style>
|
| 285 |
+
|
| 286 |
+
<a href="https://dashboard.pruna.ai/login?utm_source=huggingface&utm_medium=model_card&utm_campaign=hf_traffic" class="model-button">Try our models</a>
|