Velaura AI, a startup focused on chip design for data centers, has crossed the $1 billion valuation threshold after closing a $110 million Series A round led by Seligman Ventures, with participation f…
A developer built a team of three AI agents using the Hermes Bot Mode desktop plugin, creating a Researcher → Risk Analyst → Thesis Editor pipeline that hands off work automatically. The test demonstr…
Data Center Knowledge's most-read hardware stories from July through mid-August 2026 highlight an infrastructure-first phase of AI buildouts, with energy, minerals, and networking taking center stage.…
Modders have upgraded Nvidia's older RTX 3080 graphics cards from 10GB to 20GB of GDDR6X memory, making them a cheaper alternative for modern gaming and AI workloads, with prices around $450 on Chines…
A user on the Level1Techs forum asks whether the Ryzen AI Max 392 processor supports VFIO passthrough with vGPU splitting for running a Windows VM alongside a Linux host. The forum response indicates …
The vLLM ecosystem is shifting from batch-oriented inference to disaggregated prefill/decode serving to handle agentic traffic, a theme central to the first vLLM Conference scheduled for August 24-26,…
A Linux user seeking GPU upgrade recommendations for two machines, one with an RX580 4GB and another with an ancient 1GB card, has received suggestions from forum members favoring Intel Arc GPUs for t…
A new native installer brings vLLM and ROCm 7.15 (TheRock) to AMD Radeon RX 6000 (RDNA2) GPUs on Windows 11, achieving 26 TFLOPS FP16 and 58-62 tok/s on Qwen3.5-4B, verified on an RX 6750 XT 12 GB. Th…
Chinese cloud provider UCloud has deployed servers running Hygon's Tianxi series AI accelerators on its public-cloud platform, marking the first large-scale public-cloud rollout of the domestic chip s…
Linux Mint users report that firmware version 2.29 of the amdgpu DMCUB firmware breaks display output on AMD Radeon RX 6800 XT (Sienna Cichlid) GPUs, causing a half-black screen, while reverting to ve…
Modular Inc. announced at ModCon that Qualcomm Technologies' data center AI accelerators, starting with the Qualcomm Cloud AI 100 and extending to Qualcomm Dragonfly AI 200 and future generations, are…
AMD's ROCm blog published a walkthrough for running verl's asynchronous reinforcement learning examples on AMD Instinct MI355X GPUs, covering GRPO on Qwen2.5-VL-7B-Instruct with the Geometry3k dataset…
AMD published its 2025-26 Corporate Responsibility Report, highlighting progress on 2020-2025 goals and setting a new 2030 target to improve rack-level AI energy efficiency 20x from 2024 to 2030. The …
A research paper, “GPU Offload in Rust: Portable, Safe, and Fast” (arXiv:2608.13759), introduces GPU programming support directly into rustc, allowing developers to write GPU kernels in safe Rust with…
AMD's MI250 GPU, based on the MI200 (M) ASIC, can be adapted to PCIe using an OAM card, as demonstrated by a user who successfully enumerated the adapter with AMD's amdfwflash tool after setting the i…
Riot Platforms Inc. shares rose 4.7% to $19.91 after JPMorgan raised its price target to $22, citing the Bitcoin miner's $9.1 billion data center agreement reportedly involving Anthropic. The 20-year …
Intel has launched the Xeon 600 series workstation processors, based on Granite Rapids silicon, offering up to 86 CPU cores, a 43% increase over the previous generation, and marking the first new work…
KTransformers 0.7, the heterogeneous LLM inference and fine-tuning framework, is now available with full AVX-512 support for LoRA fine-tuning that no longer requires Advanced Matrix Extensions (AMX), …
AMD has surpassed its AI productivity target, achieving a 30 percent overall productivity boost in software development within a year, up from the 25 percent goal set for two to three years. The compa…
Best Buy's 60th Anniversary sale offers up to $1,350 off on the latest 2026 Copilot+ PCs and gaming laptops, with prices starting at $199.99. The discounts include models powered by AMD's Ryzen AI 400…