Specifications

Overview

The NVIDIA RTX PRO 6000 Blackwell is the flagship professional workstation GPU built on the Blackwell architecture, delivering 125 TFLOPS of FP32 compute from 24,064 CUDA cores paired with a massive 96 GB of GDDR7 ECC memory. With 1.8 TB/s of memory bandwidth over a 512-bit bus, it accelerates the largest LLM inference workloads — running 70B-parameter models on a single GPU — as well as QLoRA fine-tuning, video diffusion models, and professional visualization. PCIe 5.0 x16 interface, four DisplayPort 2.1b outputs, and sixth-generation NVDEC / ninth-generation NVENC engines make it ideal for AI development, data science, content creation, and enterprise deployment. Combined with NVIDIA AI Enterprise and the NemoClaw autonomous agent toolkit, the RTX PRO 6000 Blackwell is the definitive workstation GPU for 2026.