Instructions to use replicate/megablocks with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Kernels
How to use replicate/megablocks with Kernels:
# !pip install kernels from kernels import get_kernel kernel = get_kernel("replicate/megablocks") - Notebooks
- Google Colab
- Kaggle
Download torch-ext/torch_binding.h from replicate/megablocks: direct link, hf CLI and curl.
- Browser
- Download file 220 Bytes
-
https://huggingface.co/replicate/megablocks/resolve/main/torch-ext/torch_binding.h
- Command line
-
hf download hf://replicate/megablocks/torch-ext/torch_binding.h
-
curl -L -o torch_binding.h https://huggingface.co/replicate/megablocks/resolve/main/torch-ext/torch_binding.h
220 Bytes
| torch::Tensor exclusive_cumsum_wrapper(torch::Tensor x, int64_t dim, torch::Tensor out); | |
| // torch::Tensor inclusive_cumsum_wrapper(torch::Tensor x, int64_t dim, torch::Tensor out); |