| license: other | |
| library_name: pytorch | |
| pipeline_tag: image-to-video | |
| tags: [longlive-2.0, robotics, autoregressive] | |
| # LongLive 2.0 Robot-M | |
| Autoregressive robot-video generation weights. | |
| M is trained on a mixture of short and medium-length sequences. | |
| S and M describe training sequence length, not model size. | |
| ## Download | |
| ```bash | |
| hf download Efficient-Large-Model/LongLive2.0-Robot-M --local-dir ./weights/LongLive2.0-Robot-M | |
| ``` | |
| ## Inference | |
| Instantiate the matching LongLive 2.0 generator, then load the FP32 weights: | |
| ```python | |
| import torch | |
| state_dict = torch.load("model.pt", map_location="cpu", mmap=True, weights_only=True) | |
| model.load_state_dict(state_dict, strict=True) | |
| ``` | |
| Run the loading example from the downloaded directory. The matching inference | |
| runtime, VAE and text encoder are required separately; this is not a standalone | |
| pipeline. Optimizer and other training states are not included. | |
| Original upstream model terms continue to apply. | |