- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
oneAPI 2023.2 release is AVAILABLE NOW!
Intel® oneAPI Base Toolkit
- Developers now experience simplified migration from CUDA* to performant SYCL* as demonstrated by popular applications including AI deep learning, cryptography, scientific simulation, imaging, and other application areas showing equivalent performance before and after migration. Intel tools now support additional CUDA APIs, the latest version of CUDA, and FP64 for broader migration coverage.
- Developers now enjoy the freedom and productivity of CPU and GPU multi-architecture software development to Python with easy drop-in replacements for NumPy and cuPy functions plus Numba compiler support. (Beta in v2023.2, 2024.0 Production release in Q4)
- New performance and efficiency optimizations deliver the full value of Intel Xeon and data center GPU platforms. Highlights include new Intel® Open Image Denoise GPU support and major SYCL compilation and math library performance improvements on Intel® Data Center GPU Max Series.
Intel® oneAPI HPC Toolkit
- Developers will now see improved performance on Intel® Max and Flex Series GPU from drastically reduced latency on kernel launches from Intel® oneMKL.
- Fortran developers can now use DO CONCURRENT reduction to significantly improve the performance of reduction operations. And there is an even faster path to correct code by finding errors related to uninitialized variables using the LLVM memory sanitizer. Ongoing enhancements for OpenMP 5.2 compliance.
- Developers will experience higher productivity creating correct performant code with improved interoperability of OpenMP* and SYCL for CPU and GPU offload from Intel® oneAPI DPC++/C++ Compiler, along with improved debug info.
Intel® oneAPI Rendering Toolkit
- In addition to CPU support, Intel® Open Image Denoise adds GPU support with SYCL (Intel® oneAPI DPC++/C++ Compiler) for Intel® Arc, Flex and Max GPUs.
- It is also the first Rendering Toolkit component to offer cross-vendor support as it works on Intel, Nvidia, and AMD GPUs helping drive the oneAPI mission.
- Intel® Open Image Denoiseallows ISVs, studios, & researchers to provide choice for their end users most efficient computing of AI based image enhancement. They can use it on a wide range of supported hardware.
Download Intel® oneAPI Toolkits TODAY!
https://www.intel.com/content/www/us/en/developer/tools/oneapi/toolkits.html
Intel oneAPI Base and HPC Toolkits for macOS on x86 are now deprecated and will be discontinued in the oneAPI 2024.0 release. Several Intel-led open source developer tool projects will continue supporting macOS on Apple Silicon including oneAPI Threading Building Blocks (oneTBB) and Intel® Implicit SPMD Program Compiler and we welcome the opportunity to work with contributors to expand support to additional tools in the future.
Link Copied
0 Replies
Reply
Topic Options
- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page