LattePanda Sigma 2 SBC features Intel Panther Lake and 180 TOPS AI performance
Key Takeaway
The LattePanda Sigma 2 is a high-performance x86 single-board computer designed for local AI inference and robotics.
Summary
LattePanda has announced the Sigma 2, an x86 single-board computer powered by the Intel Core Ultra X7 358H Panther Lake processor. The board is optimized for local AI workloads, offering a combined theoretical peak of 180 TOPS across its CPU, NPU, and integrated GPU. It supports Ubuntu 24.04 LTS and comes pre-installed with AI frameworks and models.
Technical specifications of Sigma 2
- Processor: Intel Core Ultra X7 358H (16C/16T: 4P + 8E + 4LPE) with Intel Arc B390 graphics.
- AI Performance: 180 TOPS total (122 TOPS GPU, 50 TOPS NPU, 8 TOPS CPU) at INT8 precision.
- Memory: 64GB LPDDR5X-8533, allowing up to 59GB allocation for AI/graphics.
- Connectivity: Dual 10GbE RJ45 ports, two Thunderbolt 4 ports, and an OCuLink connector.
- Expansion: Three M.2 slots (PCIe 5.0 x2, PCIe 4.0 x4, PCIe 4.0 x1) and an onboard RP2350 microcontroller for hardware I/O.
- Software: Supports Ubuntu 24.04 LTS; includes OpenVINO, Ollama, and llama.cpp support.
Why it matters
The inclusion of pre-installed AI models and drivers for Ubuntu makes this a turnkey solution for developers working on edge AI and robotics.
Original reference: linuxgizmos.com →