Instructions to use replicate/flashinfer-draft with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Kernels
How to use replicate/flashinfer-draft with Kernels:
# !pip install kernels from kernels import get_kernel kernel = get_kernel("replicate/flashinfer-draft") - Notebooks
- Google Colab
- Kaggle
|
Download README.md from replicate/flashinfer-draft: direct link, hf CLI and curl.
- Browser
- Download file 710 Bytes
-
https://huggingface.co/replicate/flashinfer-draft/resolve/main/README.md
- Command line
-
hf download hf://replicate/flashinfer-draft/README.md
-
curl -L -o README.md https://huggingface.co/replicate/flashinfer-draft/resolve/main/README.md
710 Bytes
| license: apache-2.0 | |
| tags: | |
| - kernels | |
| > [!CAUTION] | |
| > Starting from September 13, 2026, we will be removing the "model" type repositories of kernels (e.g., kernels-community/flash-attn3). Make sure you're using a latest version of kernels. If you face any disruption, please report them here: https://github.com/huggingface/kernels/issues/new. | |
| This kernel is a work in progress and requires more work to correctly add all of the FlashInfer kernels. | |
| Please see the [generate-source.md](generate-source.md) for instructions on how to generate the source files that are contained in this kernel. | |
| ### Run `gelu_and_mul` operation | |
| ```bash | |
| nix develop -L .#test --command python tests/simple_test.py | |
| ``` |