Axelera AI, a pioneer in AI hardware acceleration, has unveiled the Metis M.2 Max the latest addition to its well-regarded Metis AI processor unit (AIPU) family. The compact M.2 module delivers PCIe-level performance, targeting the most compute-intensive AI inference tasks at the edge, including advanced large language models (LLMs) and vision transformer networks.
Taking the original Metis M.2 to the next level, the Metis M.2 Max doubles the memory bandwidth, comes in a slimmer design, includes smarter thermal management, and boosts security features. The module supports up to 16 GB of memory and comes in both standard (-20°C to +70°C) and extended (-40°C to +85°C) operating temperature ranges, making it suitable for demanding environments such as industrial manufacturing, retail, security, healthcare, and public safety.
AI Authority Trend: Righ Partners with Synaptics for Smart Home Agentic AI Applications
Fabrizio del Maffeo, CEO of Axelera AI, emphasized the company’s commitment to performance and affordability: “We continue to set the price-performance ratio benchmark for the AI accelerator market. Our goal is to make it possible for our customers to deploy transformative edge AI applications at scale. Metis M.2 Max and our original Metis M.2 card enable customers to choose the solution that best fits their requirements – from computer vision to large language models to the latest of transformer models. We are also adding enhanced security to better address the growing threat landscape at the edge.”
Metis M.2 Max leverages Axelera AI’s proven Metis AIPU architecture to deliver high performance and efficiency for demanding models such as GenAI, LLMs, and vision-language models (VLMs). The module achieves a 33% performance improvement in convolutional neural networks (CNNs) and doubles token processing speed for LLMs and VLMs, all while maintaining a typical average power consumption of just 6.5W.
AI Authority Trend: HoundDog.ai Launches Advanced Privacy-Focused Code Scanner for AI Applications
Security is a top priority at the edge, and Metis M.2 Max addresses this with enhanced firmware integrity protections. A securely provisioned Root of Trust ensures secure boot and upgrade capabilities, operating transparently to verify that only authenticated Axelera-issued firmware runs on devices.
The Metis M.2 Max comes as a 2280 M.2 M-key module and is available either on its own or with an optional low-profile heatsink. The new design reduces card height by 27%, allowing more flexible deployment configurations. Additionally, onboard power probes enable automatic performance adjustments for power- or thermal-constrained environments.
Axelera AI also supports developers through its Voyager SDK, which simplifies integration, unlocks the full potential of Metis AIPUs, and streamlines deployment of both proprietary and industry-standard models. With Metis M.2 Max, Axelera AI reinforces its leadership in high-performance, secure, and energy-efficient edge AI solutions.
AI Authority Trend: Firebolt Unveils FireScale: Redefining Analytics Benchmarks for AI Applications
To share your insights, please write to us at sudipto@intentamplify.com
