-
AMD Emerges to Challenge Nvidia on MLPerf Training
The MLPerf 5.0 training benchmarks reveal first scores for AMD’s GPUs and introduce a Llama 3.1-405B test. Nvidia and CoreWeave show Hopper and Blackwell scaling. continue reading
-
Xiaomi Xring O1 Challenges Flagship Smartphone Chips
Employing Arm-licensed IP, Xiaomi’s first self-developed smartphone processor rivals flagship chips from Apple, MediaTek, and Qualcomm but lacks an integrated 5G modem. continue reading
-
Nvidia Offers NVLink a la Carte
Nvidia’s NVLink Fusion program allows other companies to integrate its high-speed interconnect into their chips, enabling direct connections to Nvidia GPUs. Among the first licensees and reentering the data-center server market, Qualcomm is adding NVLink to custom CPUs. continue reading
-
PyXL Processor Directly Runs Python Bytecode
PyXL reduces Python execution latency and improves determinism compared with conventional CPUs, improving the language’s suitability for embedded designs. However, it’s not the first processor to directly run a high-level language. continue reading
-
Semidynamics Unveils Customizable Cervell AI Accelerator
Ranging from 8 to 128 INT8 TOPS, the licensable design uses RISC-V technology to give customers more control and customization than black-box NPUs offer. continue reading
-
InferX Tackles the GPU Cold-Start Problem
Startup InferX’s software quickly and securely loads and unloads AI models on Nvidia GPUs, reducing idle time versus dedicating a GPU to a model. continue reading
Popular
Sponsored Content and Plugs
- Slash Server Costs Without Hardware or Software Changes
- MEXT Helps IT Leaders Find the Sweet Spot
- Chips Act Backs Chiplets
- Arteris Expands NoC Offerings for AI Accelerators
- Plug: The SHD Group is Developing a Complimentary Edge-AI Processor and Ecosystem Report
Other Sites You Might Like
Read More
Altera AMD Ampere Apple Arm auto Broadcom Ceva Computex CPU data center DPU edge AI embedded Epyc FPGA Google GPU Imagination Immortalis Intel interconnect Marvell MCU MediaTek memory Meta Microchip Microsoft MLPerf NPU (AI accelerator) Nvidia NXP PC process tech Qualcomm RISC-V SambaNova Semidynamics smartphone SoftBank software Syntiant Tenstorrent Untether