Instructions to use lambda/text2bricks-360p-64f with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use lambda/text2bricks-360p-64f with Transformers:
# Load model directly from transformers import STDiT2 model = STDiT2.from_pretrained("lambda/text2bricks-360p-64f", device_map="auto") - Notebooks
- Google Colab
- Kaggle
|
Download README.md from lambda/text2bricks-360p-64f: direct link, hf CLI and curl.
- Browser
- Download file 1.22 kB
-
https://huggingface.co/lambda/text2bricks-360p-64f/resolve/main/README.md
- Command line
-
hf download hf://lambda/text2bricks-360p-64f/README.md
-
curl -L -o README.md https://huggingface.co/lambda/text2bricks-360p-64f/resolve/main/README.md
1.22 kB
metadata
license: apache-2.0
Prompt: A young man walks alone by the seaside."
Text2Bricks is a fine-tuned Open Sora model that generates toy brick-style short stop animations.
text2bricks-360p-64f is fine-tuned to generated up to 360p/64-frames outputs.
You can play with the videos created by the model in this game.
It was trained on Lambda's 1-Click Clusters in ~1,000 H100 GPU hours. See this Weights $ Biases report for details.
Extra code and data process steps can be found in this tutorial.
Usage
Use Lambda's fork of Open-Sora.
python scripts/inference.py \
configs/opensora-v1-1/inference/text2bricks-360p-64f.py \
--prompt "A young man walks alone by the seaside." \
--num-frames 64