RADX PXIe-GPUs Bring the Easy-to-Program Power of NVIDIA® Professional GPUs to PXIe for Accelerated Graphics, Video, Image, Signal & Data Processing and ML / DL / LLM Training & Inference Apps
Easy-to-Program via LabVIEW, MATLAB, Python & C/C++
with CUDA®, OpenCL® & Tensor Core Acceleration for Signal, Image & Video Processing and ML/DL/LLM Training & Inference Apps in PXIe Systems and Optimized Support for Popular Libraries & Frameworks:
- Accelerated Image, Video & Signal Processing: 1.7 to 91.1 FP32 TFLOPS for up to 100x to 200x Higher Compute Performance vs. Xilinx KU060 FPGAs and popular PXIe ECs
- NVENC / NVDEC HW Accelerated Video Encoding / Decoding Engines for Real-Time 2k/4k/8k Video Compression & Decompression at up to 100x CPU CODEC Perf Levels
- 2D/3D Graphics: Up to ~200x Higher Performance vs. Embedded Controller GPUs
- AI ML / DL / LLM Inference & Training: Up to ~200x versus Embedded Controller CPUs )
- Supports Wide Range of PXIe Chassis: GPUs for 38W, 58W and 82W/Slot PXIe Chassis
Eliminates the Need for Separate GPU Servers
- Accelerated Image, Video & Signal Processing: 1.7 to 91.1 FP32 TFLOPS for up to 100x to 200x Higher Compute Performance vs. Xilinx KU060 FPGAs and popular PXIe ECs
- NVENC / NVDEC HW Accelerated Video Encoding / Decoding Engines for Real-Time 2k/4k/8k Video Compression & Decompression at up to 100x CPU CODEC Perf Levels
- 2D/3D Graphics: Up to ~200x Higher Performance vs. Embedded Controller GPUs
- AI ML / DL / LLM Inference & Training: Up to ~200x versus Embedded Controller CPUs )
- Supports Wide Range of PXIe Chassis: GPUs for 38W, 58W and 82W/Slot PXIe Chassis