RTX A6000

The RTX A6000 is the flagship of NVIDIA's Ampere-generation RTX A-series professional lineup, the workstation counterpart to the consumer GeForce RTX 30 series. Built on the Ampere architecture, it sits two generations before the current RTX PRO Blackwell professional cards and represents the most capable single-GPU configuration of its professional generation.

The card carries 48GB of GDDR6 memory with ECC support, paired with a 384-bit memory bus delivering 768 GB/s of bandwidth. It features 10752 CUDA cores for 38.7 TFLOPS of FP32 throughput, along with 2nd-generation RT cores and 3rd-generation Tensor cores. In a dual-slot form factor on a PCIe 4.0 x16 interface and rated at 300W, it runs NVIDIA's ISV-certified professional drivers. Two cards can be linked over NVLink to pool memory for the most demanding workloads.

The RTX A6000 is designed for professional visualization, CAD/CAE, 3D rendering, real-time ray tracing, simulation, AI/ML development and inference, data science, and video production. Its large 48GB framebuffer makes it especially well suited to very large scenes and datasets, as well as running sizeable AI models locally on the workstation.