onednn
https://github.com/oneapi-src/onednn
C++
oneAPI Deep Neural Network Library (oneDNN)
Triage Issues!
When you volunteer to triage issues, you'll receive an email each day with a link to an open issue that needs help in this project. You'll also receive instructions on how to triage issues.
Triage Docs!
Receive a documented method or class from your favorite GitHub repos in your inbox every day. If you're really pro, receive undocumented methods or classes and supercharge your commit history.
C++ not yet supported2 Subscribers
Add a CodeTriage badge to onednn
Help out
- Issues
- oneDNN v3.12 release notes
- cpu: aarch64: implement forward lnorm in SVE
- aarch64: support for per_dim_0 scales and bf16 dst_dt in jit int8 matmul
- cpu, benchdnn: add reorder to/from grouped with different dts and use grouped in matmul ref
- bnorm can fail with out_of_memory on cpu
- common: memory: follow-up on large tensor dimension checks
- Copy of Upconvert fp8 weights to xf16 in Matmul in case of xf16 activations
- benchdnn: add GatedMLP driver
- rfcs: graph: add deterministic mode support
- cpu: x64: matmul: add MXFP4 weights support
- Docs
- C++ not yet supported