Instructions to use kernels-community/triton_kernels with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Kernels
How to use kernels-community/triton_kernels with Kernels:
# !pip install kernels from kernels import get_kernel kernel = get_kernel("kernels-community/triton_kernels") - Notebooks
- Google Colab
- Kaggle
Update README.md
Browse files
README.md
CHANGED
|
@@ -1,7 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
| 1 |
# triton-kernels
|
| 2 |
|
| 3 |
triton-kernels is a set of kernels that enable fast moe on different architecture. These kernels are compatible with different precision (e.g bf16, mxfp4)
|
| 4 |
|
| 5 |
|
| 6 |
Original code here https://github.com/triton-lang/triton/tree/main/python/triton_kernels
|
| 7 |
-
The current version is the following commit 7d0efaa7231661299284a603512fce4fa255e62c
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: mit
|
| 3 |
+
---
|
| 4 |
# triton-kernels
|
| 5 |
|
| 6 |
triton-kernels is a set of kernels that enable fast moe on different architecture. These kernels are compatible with different precision (e.g bf16, mxfp4)
|
| 7 |
|
| 8 |
|
| 9 |
Original code here https://github.com/triton-lang/triton/tree/main/python/triton_kernels
|
| 10 |
+
The current version is the following commit 7d0efaa7231661299284a603512fce4fa255e62c
|