Next-Level Edge AI – Now in M.2

With its new M.2 AI accelerator module based on the Kinara Ara-2, F&S brings true Generative AI performance to the edge – compact, scalable, and ready to deploy.

Delivering up to 40 TOPS of AI performance, the module enables embedded systems to run even the most demanding workloads, including large language models (LLMs), vision-language models (VLMs), and multimodal AI applications – all locally and in real time.

Designed for seamless integration, the standard M.2 form factor allows developers to upgrade existing platforms with high-performance AI acceleration via PCIe, without redesigning the system.

Key Highlights

• - Up to 40 TOPS AI performance for next-generation inference
• - Generative AI at the edge – supports LLMs, transformers & multimodal models
• - Run multiple AI models in parallel without performance degradation
• - Standard M.2 (M-Key) form factor for easy integration
• - High-speed PCIe interface for maximum data throughput
• - On-board LPDDR4 memory (up to 16 GB) for complex AI models
• - Supports TensorFlow, PyTorch, ONNX
• - Low latency & no cloud dependency

Unlock New AI Use Cases

• - Multi-camera video analytics in real time
• - AI copilots & local assistants (LLMs on the edge)
• - Advanced HMI with vision + language interaction
• - Medical imaging with AI-assisted diagnostics
• - Industrial automation with complex AI pipelines

Why F&S

• - Industrial-grade design for reliable deployment
• - Seamless integration with NXP i.MX platforms
• - Scalable AI performance across product families
• - Fast time-to-market with ready-to-use solutions
• - Development and production in Germany