The NVIDIA RTX A6000 is the flagship Ampere architecture professional workstation GPU, announced in October 2020. It doubled the memory capacity over its predecessor (Quadro RTX 6000) while introducing Ampere's enhanced Tensor Cores and RT cores to professional workflows.
The RTX A6000 uses the full GA102 die with 48GB of GDDR6 ECC memory providing 768 GB/s bandwidth. It includes 10,752 CUDA cores, 336 third-generation Tensor Cores, and 84 second-generation RT cores. The chip is manufactured on Samsung's 8nm process.
The 48GB memory capacity was groundbreaking for workstation GPUs, enabling professionals to work with datasets previously requiring datacenter hardware. This capacity supports large CAD assemblies, high-resolution video editing, complex 3D scenes, and AI model development.
NVLink support enables connecting two RTX A6000 cards with 112.5 GB/s bidirectional bandwidth for a combined 96GB memory pool. Professional certifications cover all major ISV applications. The RTX A6000 remains widely deployed and continues to be available alongside newer Ada Lovelace products.