| --- |
| base_model: PAIR/StreamingSVD |
| tags: |
| - quantized |
| - int8 |
| - image-to-video |
| --- |
| |
| # StreamingSVD — INT8 Quantized |
|
|
| This is an INT8 symmetric per-channel quantization of |
| [PAIR/StreamingSVD](https://huggingface.co/PAIR/StreamingSVD) by |
| [Picsart AI Research (PAIR)](https://huggingface.co/PAIR). |
|
|
| > **StreamingSVD: Consistent, Dynamic, and Extendable Image-Guided Long Video Generation** |
| > Roberto Henschel, Levon Khachatryan, Daniil Hayrapetyan, Hayk Poghosyan, |
| > Vahram Tadevosyan, Zhangyang Wang, Shant Navasardyan, Humphrey Shi |
| > [[arXiv 2403.14773]](https://arxiv.org/abs/2403.14773) · [[Project page]](https://streamingt2v.github.io) · [[Code]](https://github.com/Picsart-AI-Research/StreamingT2V) |
|
|
| ## Quantization |
|
|
| | | Size | Dtype | |
| |---|---|---| |
| | Original | ~12.5 GB | F32 | |
| | **This repo** | **~3.2 GB** | **INT8** | |
|
|
| Large weight tensors are quantized with symmetric per-channel INT8. |
| Norms, biases, and positional embeddings are kept in FP32 for accuracy. |
|
|
| ## Loading |
|
|
| ```python |
| from safetensors.torch import load_file |
| import torch |
| |
| tensors = load_file("model.int8.safetensors") |
| weights = {} |
| for k, v in tensors.items(): |
| if k.endswith(".scale"): |
| continue |
| scale_key = k + ".scale" |
| if scale_key in tensors: |
| scale = tensors[scale_key] |
| shape = [-1] + [1] * (v.dim() - 1) |
| weights[k] = v.float() * scale.view(shape) |
| else: |
| weights[k] = v |
| ``` |
|
|
| ## Citation |
|
|
| ```bibtex |
| @article{henschel2024streamingt2v, |
| title={StreamingT2V: Consistent, Dynamic, and Extendable Long Video Generation from Text}, |
| author={Henschel, Roberto and Khachatryan, Levon and Hayrapetyan, Daniil and Poghosyan, Hayk |
| and Tadevosyan, Vahram and Wang, Zhangyang and Navasardyan, Shant and Shi, Humphrey}, |
| journal={arXiv preprint arXiv:2403.14773}, |
| year={2024} |
| } |
| ``` |
