diff --git a/README.md b/README.md index 1038a54e3..24da9f8f9 100644 --- a/README.md +++ b/README.md @@ -14,7 +14,7 @@ Some of the key features of BitBLAS include: - Support customizing mixed-precision DNN operations for your specific scenarios via the flexible DSL (TIR Script). ## Latest News - +- 11/04/2024 🚀🚀: We've supported high performance A INT4 x W INT4/INT2 Matmul. - 10/02/2024 🚀🚀: We've added initial Flash Attention Ops and its implementation in Tilelang! Please refer to [PythonAPI](https://github.com/microsoft/BitBLAS/blob/main/docs/PythonAPI.md) and [QuickStart](https://github.com/microsoft/BitBLAS/blob/main/docs/QuickStart.md) docs and [PR #202](https://github.com/microsoft/BitBLAS/pull/202). - 08/12/2024 🚀🚀: We've improved performance for contiguous batching. To enable it, you'll need to set specific flags. For more details, please refer to [PR #133](https://github.com/microsoft/BitBLAS/pull/133). - 07/11/2024 ✨: Ladder is published and presented in OSDI'24. Please find [Ladder paper and presentation](https://www.usenix.org/conference/osdi24/presentation/wang-lei) if you are interested in the technical details of BitBLAS.