We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
1 parent db1c55d commit 6a79dc3Copy full SHA for 6a79dc3
1 file changed
advanced_source/cuda_graph_annotations_tutorial.py
@@ -20,10 +20,10 @@
20
.. grid-item-card:: :octicon:`list-unordered;1em;` Prerequisites
21
:class-card: card-prerequisites
22
23
- * PyTorch 2.0 or later
+ * PyTorch 2.12+
24
* CUDA-capable GPU
25
- * Driver/CUDA-compat >= 13.1 for full annotation support
26
- * cuda-python package
+ * Driver/CUDA-compat >= 13.1 for annotation support
+ * cuda-bindings >= 13.3.0
27
28
CUDA graphs are a powerful optimization technique that can significantly reduce
29
kernel launch overhead by capturing and replaying sequences of CUDA operations.
0 commit comments