PNY NVIDIA RTX Pro 6000 Blackwell Server Edition

Fifth-generation Tensor Cores: Deliver up to five times the performance of the previous generation for large-scale language model (LLM) inference, with support for FP4 accuracy to accelerate AI model processing and reduce memory usage. Fourth-generation ray tracing cores: Double the ray-triangle intersection rate compared to the previous generation to create photorealistic and physically faithful scenes as well as immersive 3D designs with RTX Mega Geometry technology, which can process up to 100x more triangles in ray tracing. Next-generation video engines: Accelerate multimodal AI inference, optimize video production, and boost streaming workflows with real-time AI processing. The RTX PRO 6000 Blackwell features four (4) ninth-generation NVENC engines and four (4) sixth-generation NVDEC motors, supporting 4:2:2 encoding and decoding. GDDR7 memory: New and improved GDDR7 memory dramatically increases bandwidth and capacity, allowing your applications to run faster and process larger, more complex data sets. With 96 GB of GPU memory and 1.6 TB bandwidth, businesses can achieve low-latency inferences with larger models and render large-scale 3D scenes. Universal MIG: Split a single RTX PRO 6000 Blackwell server into up to four fully isolated instances, each with dedicated resources, enabling graphics and AI workloads to run simultaneously, optimized GPU utilization, and secure isolation of different applications or users.