TechNexion Telos-AI4000 Series is a compact, high‑performance AI acceleration module that elevates Telos-Vision and Telos-Core and delivers 40 TOPS of AI performance, enabling low‑latency, energy‑efficient AI inference for generative AI and demanding edge computing applications.
Powered by NXP Ara 240, the Telos-AI4000 Series delivers up to 40 TOPS. It balances high throughput with efficiency for edge and generative AI workloads.
Automatically detected by Telos Vision and Telos Core devices. Add 40 TOPS of AI performance within seconds.
Compatible with Tensorflow PyTorch, ONNX, Caffe, and more. Deplay YOLOv8 and LLAaMA 2 models in a heartbeat.
Supports Vision Language Models (VLMs) and Vision Language Actions (VLAs), Transformer Models and Convolutional Neural Networks (CNNs).
Ultra low Latency : ResNet50 in 2 ms provided over a superfast PCIe interface integrated with Telos Vision DMA support.
Up to 16GB low power double data rate (LPDDR4) memory provides scalability to match your AI Model needs.
Telos-AI4000 Series comes with a 10 year longevity commitment and pin-compatible upgrades.
Software Integration Partners can run their software on any Telos-Vision or Telos-Core product locally.
