• NVIDIA RTX A1000 Embedded GPU based on Ampere architecture
• 2048 CUDA cores, 16 RT cores and 64 Tensor cores, 4GB GDDR6 memory
• 6.66 TFLOPS peak FP32 performance
• PCIe Gen 4 x8 interface

• NVIDIA RTX A2000 Embedded GPU based on Ampere architecture
• 2560 CUDA cores, 20 RT cores and 80 Tensor cores, 8GB GDDR6 memory
• 8.25 TFLOPS peak FP32 performance
• Support Error Correction Code(ECC)
• Support CUDA Compute version 8.6, OpenGL 4.6, Vulkan 1.2, DirectX 12 Ultimate and Shader Model 7.0

• NVIDIA RTX A4500 Embedded GPU based on Ampere architecture
• 5888 CUDA cores, 80 RT cores and 184 Tensor cores, 16GB GDDR6 memory
• 18.55 TFLOPS peak FP32 performance
• Support Error Correction Code(ECC)
• Support CUDA Compute version 8.6, OpenGL 4.6, Vulkan 1.2, DirectX 12 Ultimate and Shader Model 7.0

• NVIDIA RTX A500 Embedded GPU based on Ampere architecture
• 2048 CUDA cores, 16 RT cores and 64 Tensor cores, 4GB GDDR6 memory
• PCIe Gen 4 x4 interface
• 7.28 TFLOPS peak FP32 performance