A New Era of Accelerated Computing
Benchmarks: Intel® Xeon® CPU Max Series & Intel® Data Center GPU Max Series
Intel® oneAPI tools and optimized AI frameworks deliver leading performance on popular deep learning and molecular dynamics benchmarks.
Performance Leap in Inference & Training
oneAPI tools benchmark test results for MLPerf* DeepCAM, commonly used in HPC data centers for detection of hurricanes and atmospheric rivers in climate data.
Deep Learning Inference & Training Boost
Hardware acceleration with Intel® Advanced Matrix Extensions (Intel® AMX) and support for int8 (inference) and bfloat16 (training/inference) datatypes provide significant performance gains for AI and deep learning inference and training workloads compared to previous generation Intel® Xeon® Scalable processors and competition. TensorFlow* and PyTorch* frameworks running on 4th generation Intel Xeon Scalable Processors deliver leading AI performance with Intel AMX support and extended optimization capabilities enabled through the Intel® oneAPI Deep Neural Network Library (oneDNN).
Speeding Exascale Material Discovery
oneAPI tools benchmark test results for the Liquid Crystal workload in LAMMPS, a popular molecular dynamics code for life science and materials research.
Life & Material Science Speed-Ups
This benchmark was run on a combination of Intel® Xeon CPU Max Series and Intel® Data Center GPU Max Series using both DDR and high-bandwidth memory. Many of the brain-mapping codes are written in C++, so the Intel® oneAPI DPC++/C++ Compiler—the oneAPI implementation of C++ with SYCL—was used to expose parallelism and vectorization to effectively use the GPUs. OpenMP* was also used in the compiler for parallelism in other portions of the code along with the Intel® oneAPI Math Kernel Library (oneMKL).
oneAPI: Putting the Needs of Developers First
oneAPI is an open, cross-architecture programming model that frees developers to use a single code base across multiple architectures. The result is accelerated compute without vendor lock-in.
Intel’s 2023 oneAPI & AI Tools Deliver Multiarchitecture Performance
The latest oneAPI and AI 2023 tools continue to empower developers with performance and productivity, delivering optimized support for Intel’s upcoming portfolio of CPU and GPU architectures and advanced capabilities.
Intel oneAPI 2023 Release: Preview the Tools [1:01:48]
The Multiarchitecture, Multivendor Path to Accelerated Computing [white paper]
Compare CPUs, GPUs & FPGAs for oneAPI Workloads [user guide]
Industry Showcase
See the latest collaborations, use cases, and success stories facilitated by oneAPI.
29 Universities & Counting
Across the globe, the world’s most prestigious universities are now part of the oneAPI Academic Community as Centers of Excellence.
Explore all oneAPI Centers of Excellence
Discover More
Dev-to-Dev Insights & POV
Quicken Text Classification Up to 16.47x
Xinyu Ye, Haihao Shen, Moshe Wasserblat, and Ofir Zafrir, AI frameworks/research engineers, Intel
See how Intel® Neural Compressor distills and quantizes a BERT-Mini model to accelerate inference and maintain the accuracy.
SYCL*–How the Standard Got Started
Robert Mueller-Albrecht, oneAPI Library Tools Marketing, Intel
Get a brief history of SYCL and why it was (and is) the right solution for developers.
Two Keys for an Efficient Data Science Workstation
Dr. Deb Bharadwaj, director, AI & Machine Learning Product Management, Intel
Learn the two things Intel and Anaconda* experts agree are necessary when designing an efficient data science workstation.
An Introduction to Accelerator & Parallel Programming
Tony Mongkolsmai, software architect and technical evangelist, Intel
Learn the basics, including simple ways you can run a program on multiple CPUs or an accelerator.
产品和性能信息
性能因用途、配置和其他因素而异。请访问 www.Intel.cn/PerformanceIndex 了解更多信息。
英特尔® 技术的功能和优势取决于系统配置，并且可能需要启用硬件、软件或服务才能激活。实际性能可能因系统配置的不同而有所差异。没有任何产品或组件能够做到绝对安全。请咨询您的系统制造商或零售商，或者访问 Intel.cn 了解更多信息。