metadata
license: apache-2.0
base_model: Wan-AI/Wan2.1-T2V-1.3B
pipeline_tag: video-to-video
library_name: pytorch
tags:
- video-generation
- camera-control
- wan
- rerope
ReRoPE V2V checkpoint
Inference-only V2V weights for ReRoPE, based on Wan2.1-T2V-1.3B.
Files
| File | Parameters | Dtype | Size | SHA-256 |
|---|---|---|---|---|
rerope_v2v_wan21_1.3b.safetensors |
1,418,996,800 | bfloat16 | 2,838,077,456 bytes | 31f74e89840b9ff213a4adf22c0e31d69ec5e6edacf793b00c51c475767d6ee8 |
The file contains 825 tensors and only the inference state dictionary. Optimizer state, trainer state, datasets, and local paths are not included.
Usage
Use these weights with the inference code at github.com/Sisyphe-lee/ReRoPE_.
The Wan2.1 base model is required separately. See the GitHub README for the expected directory layout and V2V command.
Scope
This initial model release contains one V2V checkpoint. The repository includes I2V inference code, but no I2V ReRoPE checkpoint is released.