Cute DSL: How to generate cubin for a different GPU arch #2557
Unanswered
victoroliv2
asked this question in
Q&A
Replies: 1 comment 1 reply
-
|
Not yet but this ahead of time compilation model is something we are actively working on. We will not exit CuTe DSL beta until we ship this. |
Beta Was this translation helpful? Give feedback.
1 reply
-
|
@thakkarV are there news on this? Any timeline? |
Beta Was this translation helpful? Give feedback.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
I’m looking into using Cute DSL to write some matmul kernels. I see that @cute.kernel and @cute.jit decorators can be used to build/run kernels. I’m looking into a way to generate cubins/sass from Python for different arquitectures and embed that into an existing C++ application for execution. Is there a way to do that?
Beta Was this translation helpful? Give feedback.
All reactions