Instructions to use replicate/fp8-fbgemm with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Kernels
How to use replicate/fp8-fbgemm with Kernels:
# !pip install kernels from kernels import get_kernel # a version (or an explicit revision) is required; see the "Files and versions" tab for the available ones kernel = get_kernel("replicate/fp8-fbgemm", version=1) - Notebooks
- Google Colab
- Kaggle
Download build/torch-cpu/_ops.py from replicate/fp8-fbgemm: direct link, hf CLI and curl.
- Browser
- Download file 193 Bytes
-
https://huggingface.co/replicate/fp8-fbgemm/resolve/main/build/torch-cpu/_ops.py
- Command line
-
hf download hf://replicate/fp8-fbgemm/build/torch-cpu/_ops.py
-
curl -L -o _ops.py https://huggingface.co/replicate/fp8-fbgemm/resolve/main/build/torch-cpu/_ops.py
193 Bytes
| import torch | |
| ops = torch.ops._fp8_fbgemm_5f3c84f_dirty | |
| def add_op_namespace_prefix(op_name: str): | |
| """ | |
| Prefix op by namespace. | |
| """ | |
| return f"_fp8_fbgemm_5f3c84f_dirty::{op_name}" |