| Name | Modified | Size | Downloads / Week |
|---|---|---|---|
| Parent folder | |||
| README.md | 2026-04-17 | 649 Bytes | |
| v3.11.3 source code.tar.gz | 2026-04-17 | 13.8 MB | |
| v3.11.3 source code.zip | 2026-04-17 | 17.2 MB | |
| Totals: 3 Items | 31.0 MB | 0 | |
This is a patch release containing the following changes to v3.11.2:
- Fixed undefined behavior in matmul implementation on Intel64/AMD64 CPUs (bd117e4438b240df7c8162ece744685672686f18)
- Fixed performance regression in
f32reorder on Intel64/AMD64 CPUs (a4acece010b8618d06a09feea0c5319da1ea3040) - Fixed a
SEGFAULTin binary primitive with large sizes on Intel GPUs (157cba5eb115a17927b262bad7f9a451233a8774, 1a8bc113ae46a31475d3f71cc668c03f34dcec24) - Fixed performance regression in
f32convolution with small number of input channels on processors with Intel AVX-512 instruction set support (e1f4a614a0b27b8f0b5d7deadd56fc9b6864bbeb)