Skip to content

cudnn FE 1.6.1 release

Choose a tag to compare

View all tags
@Anerudhan Anerudhan released this 20 Aug 04:14
· 59 commits to main since this release
v1.6.1
2533f5e
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.

Bug fix

  • Fixed an issue where custom dropout mask was not correctly applied.
  • Added -fvisibility=hidden for the pip wheels generated to avoid symbol conflicts with other modules that use cudnn frontend.
  • Fixed an issue in sdpa operation which when deserialized will lead to numerical mismatches.
  • Fixed an issue in sdpa fp8 fprop operation (in inference mode).

Samples

  • Added a new sample to showcase how a custom dropout mask can be applied to a sdpa operation.
  • Added a sample to shocase convolutions on large (c * d * h * w > 2 **31) tensors.
Assets 2
Loading