Instructions to use replicate/flash-attn2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Kernels
How to use replicate/flash-attn2 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/flash-attn2", version=1) - Notebooks
- Google Colab
- Kaggle
Download .gitignore from replicate/flash-attn2: direct link, hf CLI and curl.
- Browser
- Download file 164 Bytes
-
https://huggingface.co/replicate/flash-attn2/resolve/main/.gitignore
- Command line
-
hf download hf://replicate/flash-attn2/.gitignore
-
curl -L -o .gitignore https://huggingface.co/replicate/flash-attn2/resolve/main/.gitignore
164 Bytes
| .bak | |
| __pycache__ | |
| build-ext | |
| cmake | |
| result | |
| CMakeLists.txt | |
| setup.py | |
| pyproject.toml | |
| .venv | |
| torch-ext/registration.h | |
| torch-ext/flash_attn/*.so | |
| torch-ext/flash_attn/_ops.py |