SiFive launches 2nd generation RISC-V intelligence family for Artificial Intelligence workloads

SiFive introduced its 2nd Generation Intelligence family of RISC-V products to accelerate Artificial Intelligence workloads across thousands of potential applications. The lineup includes two new chips for the far edge and three upgraded designs with enhanced scalar, vector and matrix processing.

SiFive announced its 2nd Generation Intelligence family, a set of five RISC-V-based products designed to accelerate Artificial Intelligence workloads. The new lineup comprises two all-new parts, the X160 Gen 2 and X180 Gen 2, alongside upgraded X280 Gen 2, X390 Gen 2 and XM Gen 2 models. The company framed the release as an expansion of its RISC-V Artificial Intelligence IP portfolio to cover a broad set of application needs.

All members of the family include enhanced scalar and vector processing capabilities tailored to modern Artificial Intelligence workloads. The XM Gen 2 is highlighted for adding or emphasizing matrix processing capability in addition to scalar and vector units. SiFive positions these architectural elements as responses to the mixes of compute primitives used by contemporary machine learning and inference tasks, aiming to better match processing types to workload requirements.

SiFive singled out the X160 Gen 2 and X180 Gen 2 as targets for far edge compute and internet of things applications, noting their focus on high efficiency within a compact footprint. The company emphasized use in deeply embedded edge environments that are tightly constrained by power and area. Industries explicitly cited as targets include automotive, autonomous robotics, industrial automation and smart IoT. The remaining three products are presented as upgraded offerings within the same 2nd Generation family, broadening SiFive´s product range for different deployment scenarios and application scales.

65

Impact Score

AMD and Rackspace plan dedicated AI compute rollout

AMD and Rackspace have finalized a phased deployment for dedicated AMD-based compute across Rackspace data centers. The capacity is aimed at regulated enterprise workloads, including clinical AI and large-scale inference.

Lexar tests SSD offloading for local AI models

Lexar is developing an AI-focused SSD approach designed to cut DRAM demand when running large language models on consumer PCs. Internal tests show the company’s storage offloading can load models that traditional local frameworks struggle to run with limited memory.

NVIDIA Blackwell leads MLPerf Training 6.0

NVIDIA’s latest MLPerf Training 6.0 results put Blackwell across every benchmark in the suite, including new MoE workloads. Partner systems from Microsoft Azure and CoreWeave highlighted large-cluster runs on Llama 3.1 405B and DeepSeek-V3 671B.

Contact Us

Got questions? Use the form to contact us.

Contact Form

Clicking next sends a verification code to your email. After verifying, you can enter your message.