As enterprises race to adopt generative AI and bring new services to market, the demands on data center infrastructure have never been greater. Training large language models is one challenge, but delivering LLM-powered real-time services is another. In the latest round of MLPerf industry benchmarks, Inference v4.1, NVIDIA platforms delivered leading performance across all data center tests. The first-ever submission of the upcoming NVIDIA Blackwell platform revealed up to 4x more performance than the NVIDIA H100 Tensor Core GPU on MLPerf's biggest LLM workload, Llama 2 70B, thanks to its use of a second-generation Transformer Engine and FP4 Tensor Cores.

The NVIDIA H200 Tensor Core GPU delivered outstanding results on every benchmark in the data center category - including the latest addition to the benchmark, the Mixtral 8x7B mixture of experts (MoE) LLM, which features a total of 46.7 billion parameters, with 12.9 billion parameters active per token. MoE models have gained popularity as a way to bring more versatility to LLM deployments, as they're capable of answering a wide variety of questions and performing more diverse tasks in a single deployment. They're also more efficient since they only activate a few experts per inference - meaning they deliver results much faster than dense models of a similar size.

The latest version of AMD's open-source GPU compute stack, ROCm, is due for launch soon according to a Phoronix article—chief author, Michael Larabel, has been poring over Team Red's public GitHub repositories over the past couple of days. AMD ROCm version 6.0 was released last December—bringing official support for the AMD Instinct MI300A/MI300X, alongside PyTorch improvements, expanded AI libraries, and many other upgrades and optimizations. The v6.0 milestone placed Team Red in a more competitive position next to NVIDIA's very mature CUDA software layer. A mid-February 2024 update added support for Radeon PRO W7800 and RX 7900 GRE GPUs, as well as ONNX Runtime.

Larabel believes that "ROCm 6.1" is in for an imminent release, given his tracking of increased activity on publicly visible developer platforms: "For MIPOpen 3.1 with ROCm 6.1 there's been many additions including new solvers, an AI-based parameter prediction model for the conv_hip_igemm_group_fwd_xdlops solver, numerous fixes, and other updates. AMD MIGraphX will see an important update with ROCm 6.1. For the next ROCm release, MIGraphX 2.9 brings FP8 support, support for more operators, documentation examples for Whisper / Llama-2 / Stable Diffusion 2.1, new ONNX examples, BLAS auto-tuning for GEMMs, and initial code for MIGraphX running on Microsoft Windows." The change-logs/documentation updates also point to several HIPIFY for ROCm 6.1 improvements—including the addition of CUDA 12.3.2 support.

Quantum coherence at room temperature has been achieved, thanks to the efforts of Associate Professor Nobuhiro Yanai and his research team at Kyushu University's Faculty of Engineering. Additional credit goes to Associate Professor Kiyoshi Miyata (also of Kyushu University) and Professor Yasuhiro Kobori of Kobe University, all in Japan. Their scientific experiments have led to an ideal set of conditions where it is "crucial to generate quantum spin coherence in the quintet sublevels by microwave manipulation at room temperature." A quantum system requires operation in a stable state over a certain period of time, free of environmental interference.

Kobori-san has disclosed multi-department research results in a very elaborate document: "This is the first room-temperature quantum coherence of entangled quintets." The certain period of time mentioned above was only measures in nanoseconds, so more experimental work and further refinement will be carried out to prolong harmonious conditions. Head honco, Professor Yanai outlined some goals: "It will be possible to generate quintet multiexciton state qubits more efficiently in the future by searching for guest molecules that can induce more such suppressed motions and by developing suitable MOF structures...This can open doors to room-temperature molecular quantum computing based on multiple quantum gate control and quantum sensing of various target compounds."

After going through a month-long beta stage, Chrome 17 has now been promoted and is available as a Stable release for Windows, Mac OS and Linux. Chrome 17 (build 17.0.963.46) features new Extension APIs, updated Omnibox Prerendering, download scanning protection, various tweaks and fixes for about 20 vulnerabilities.

Chrome 17 Stable can be downloaded via this page.

Keeping the Chrome crowd well-stocked on new releases, Google has made available a couple of versions of its browser, the 16.0.912.75 (Stable channel) and 17.0.963.26 (Beta channel).

The new Chrome 16 build is a security update which comes to fix three vulnerabilities rated 'High', while the first Chrome 17 release brings quite a few fresh goodies like New Extensions APIs, Omnibox Prerendering (the browser will load a website in the background while you type in the URL), enhanced download protection, and a few tweaks.

Both Chrome 16.0.912.75 and 17.0.963.26 are available for Windows, Mac OS and Linux and can be downloaded via this page.