Instructions to use replicate/quantization-bitsandbytes with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Kernels
How to use replicate/quantization-bitsandbytes with Kernels:
# !pip install kernels from kernels import get_kernel kernel = get_kernel("replicate/quantization-bitsandbytes") - Notebooks
- Google Colab
- Kaggle
Download build/torch211-cxx11-cpu-x86_64-linux/_ops.py from replicate/quantization-bitsandbytes: direct link, hf CLI and curl.
- Browser
- Download file 273 Bytes
-
https://huggingface.co/replicate/quantization-bitsandbytes/resolve/main/build/torch211-cxx11-cpu-x86_64-linux/_ops.py
- Command line
-
hf download hf://replicate/quantization-bitsandbytes/build/torch211-cxx11-cpu-x86_64-linux/_ops.py
-
curl -L -o _ops.py https://huggingface.co/replicate/quantization-bitsandbytes/resolve/main/build/torch211-cxx11-cpu-x86_64-linux/_ops.py
273 Bytes
| import torch | |
| from . import _quantization_bitsandbytes_cpu_dbc4d19 | |
| ops = torch.ops._quantization_bitsandbytes_cpu_dbc4d19 | |
| def add_op_namespace_prefix(op_name: str): | |
| """ | |
| Prefix op by namespace. | |
| """ | |
| return f"_quantization_bitsandbytes_cpu_dbc4d19::{op_name}" | |