M.2 MAX AI Inference Acceleration card

M.2 Max: Axelera® AI’s M.2 Max card, powered by Metis AIPU, unlocks LLM and VLM applications on constrained edge and embedded devices. 

Key Benefits:

• M.2 Max featuring Metis AIPU unlocks high performance for LLMs and VLMs by maximizing memory performance on small footprint devices and with power constraints.
• The card can run computer vision inference tasks on multiple cameras using multiple cascaded or parallel models.
• The M.2 Max card is available in memory sizes up to 16 GB* and is offered standalone or with a cooling solution providing, flexibility across a range of application needs.
• It is available in standard operating temperature range (-20°C to +70°C) as well as extended operating temperature range (-40°C to +85°C).
• Firmware integrity protection is provided through secure boot and secure upgrade features built on a hardware Root-of-Trust in the Metis AIPU.
• A wide range of end-to-end AI pipelines and models are available out of the box.
• Hassle free evaluation and software integration thanks to Voyager® SDK.
• Uncompromised prediction accuracy thanks to advanced quantization tools.

*The M.2 Max card is also available in additional memory configurations for volume orders. Contact us to discuss specifications.

System Requirements

Metis M.2 Max Card has the following system integration requirements. Our team actively tests integration with different systems from vendors such as Dell, Lenovo, Advantech and Aetina.

• Host Interface: PCIe Gen. 3.0 x4
• Mechanical form factor: M.2 2280 M-key requires H5.6.
• Power Rating: Compliant with PCI-SIG’s M.2 Specification revision 4.0 (11.55 W average power, 23.1 W peak power).
• Host CPU: Intel® Core™ Processors, AMD Ryzen™ Processors, Arm64 (aarch64) based processors.
• Host Operating System:
  • Native Linux support is tested on Ubuntu 22.04
  • A Docker Guide is available to evaluate or develop with other Linux distributions.