NVIDIA RTX PRO 6000 Blackwell 96GB Workstation Graphics Card

The NVIDIA RTX PRO 6000 Blackwell is a professional workstation graphics card built on NVIDIA's Blackwell GPU architecture. Designed for AI developers, researchers, engineers, and visual effects professionals, it brings together a large memory pool, high-bandwidth interconnect, and enterprise-grade reliability into a single dual-slot card. At the heart of this card are 24,064 CUDA cores paired with fifth-generation Tensor Cores and fourth-generation RT (Ray Tracing) Cores. The Tensor Cores deliver up to 4,000 AI TOPS, enabling demanding tasks such as large language model fine-tuning, neural network training, and real-time AI inference directly on a desktop workstation. The RT Cores accelerate ray-traced rendering, making the card well-suited for photorealistic visualisation, architectural rendering, and cinematic VFX work. The card carries 96 GB of GDDR7 memory with ECC (Error Correcting Code) protection across a 512-bit memory interface, yielding 1,792 GB/s of memory bandwidth. ECC support is critical for scientific computing and simulation workloads where data integrity cannot be compromised. Multi-Instance GPU (MIG) partitioning allows the card to be divided into up to four 24 GB instances, two 48 GB instances, or used as a single 96 GB instance, providing workload isolation and flexible resource allocation. Connectivity is handled via a PCIe 5.0 x16 system interface, ensuring the card is ready for current-generation workstation platforms. Four Mini DisplayPort 2.1b outputs support up to four displays simultaneously, with resolutions up to 7680×4320 (8K) at 60 Hz on two screens or 4096×2160 at 120 Hz across four screens. Three ninth-generation NVENC and three sixth-generation NVDEC video engines support hardware-accelerated video encoding and decoding. Cooling is managed by a double-flow-through thermal solution that routes airflow through dual axial pathways to efficiently dissipate heat from the 600 W total board power envelope. The card measures 5.4 inches in height and 12 inches in length in a dual-slot, extended-height form factor. Supported graphics APIs include DirectX 12 (Shader Model 6.6), OpenGL 4.63, and Vulkan 1.4. Compute workloads are supported via CUDA 12.8, OpenCL 3.0, and DirectCompute. Key Specifications: - GPU Architecture: NVIDIA Blackwell - CUDA Cores: 24,064 - Tensor Cores: Fifth generation - Ray Tracing Cores: Fourth generation - AI Performance: 4,000 AI TOPS - GPU Memory: 96 GB GDDR7 with ECC - Memory Interface: 512-bit - Memory Bandwidth: 1,792 GB/s - System Interface: PCIe 5.0 x16 - Display Connectors: 4× Mini DisplayPort 2.1b - MIG Support: Up to 4× 24 GB / 2× 48 GB / 1× 96 GB - Total Board Power: 600 W - Thermal Solution: Double-flow-through - Form Factor: 5.4″ H × 12″ L, dual slot, extended height - Video Engines: 3× NVENC (9th gen), 3× NVDEC (6th gen) - Graphics APIs: DirectX 12, Shader Model 6.6, OpenGL 4.63, Vulkan 1.4 - Compute APIs: CUDA 12.8, OpenCL 3.0, DirectCompute