Semiconductors

Celestial AI secures $175 million in Series C funding for its Optical Compute Interconnect, promising lower latency and power consumption. Exciting advancements in optical interconnect technology!

Acacia introduces 800G DSP-based pluggable modules for metro DCI upgrades, supporting QSFP-DD and OSFP form factors compliant to OIF 800ZR.

Coherent Corp. unveils I-temp 100G ZR QSFP28 DCO transceiver for outdoor deployment in edge networks, offering expanded temperature range and low power consumption.

The U.S. Department of Commerce proposes $8.5 billion in funding for Intel's semiconductor projects, aiming to boost American innovation in AI era.

NVIDIA introduces Blackwell platform, honoring mathematician David Harold Blackwell. Leveraging six new technologies for AI training and LLM inference.

Broadcom unveils Bailly, the first 51.2 Tbps CPO switch for AI clusters, promising lower power consumption and higher efficiency. #BroadcomAI #CPO

Cerebras Systems unveils the Wafer Scale Engine 3 chip with 4 trillion transistors and 125 petaflops of AI performance, designed for next-gen AI models.

"Alphawave Semi unveils a game-changing chiplet-connectivity silicon platform utilizing TSMC's 3nm process, promising improved connectivity and computing capabilities."

Marvell and TSMC collaborate to produce 2nm semiconductors, aiming to revolutionize infrastructure with accelerated AI capabilities.

Broadcom begins sampling the first 5nm PCIe Gen 5.0/CXL2.0 and PCIe Gen 6.0/CXL3.1 retimers, offering end-to-end PCIe solutions for AI infrastructure. The company's wide-ranging portfolio includes industry-first features and benefits,...

Intel has announced plans to relaunch Altera as a standalone FPGA company, led by CEO Sandra Rivera. They also unveiled new products and services, including Agilex 9 and the FPGA...

Synopsys introduces a 1.6T Ethernet IP solution to enhance AI and data center applications, reducing time to market and improving energy efficiency.

Marvell is open sourcing software for building machine learning models on OCTEON 10 processors, enabling OEMs and operators to optimize network performance and efficiency.