PNY RTX PRO 6000 Blackwell Server Edition 96GB GDDR7 TCSRTXPRO6000SE-BLK

1. Massive 96GB GDDR7 ECC Memory for AI and HPC The RTX PRO 6000 Server Edition is equipped with 96GB of ECC-protected GDDR7 memory on a 512-bit memory interface , delivering up to 1.6 TB/s of memory bandwidth . This enormous memory capacity enables large AI models, scientific simulations, digital twins, rendering workloads, and enterprise-scale data analytics that would exceed the capabilities of traditional workstation GPUs. 2. Exceptional Blackwell AI Compute Performance Built on NVIDIA's latest Blackwell architecture , the GPU features 24,064 CUDA cores , 752 fifth-generation Tensor Cores , and 188 fourth-generation RT Cores . It delivers up to 4 PFLOPS of FP4 AI performance , 2 PFLOPS FP8 performance , and 120 TFLOPS FP32 compute , making it ideal for AI inference, model fine-tuning, engineering simulation, scientific computing, and advanced visualisation workloads. 3. Purpose-Built for Enterprise Multi-GPU Servers Unlike the workstation version, the Server Edition uses a passive cooling design optimized for rack servers and high-density deployments. It supports PCIe Gen5 x16 , Multi-Instance GPU (MIG) with up to four isolated 24GB instances, GPUDirect technologies, and configurable power from 400W to 600W , allowing enterprises to deploy multiple GPUs efficiently for AI factories, virtual workstations, rendering farms, and data-center workloads. Key Specifications Specification RTX PRO 6000 Server Edition Architecture NVIDIA Blackwell CUDA Cores 24,064 Tensor Cores 752 (5th Gen) RT Cores 188 (4th Gen) Memory 96GB GDDR7 ECC Memory Interface 512-bit Memory Bandwidth Up to 1.6 TB/s AI Performance Up to 4 PFLOPS (FP4) FP32 Performance 120 TFLOPS Interface PCIe 5.0 x16 Cooling Passive MIG Support Up to 4 Instances Power Consumption 400–600W Form Factor FHFL Dual-Slot