NVIDIA launches its brand-new RTX PRO™ 6000 Blackwell GPU
NVIDIA RTX PRO™ 6000 GPU series: 3 ranges to power up in 2025
NVIDIA revolutionizes professional computing once again with the launch of the RTX PRO™ 6000 Blackwell GPUs. Its most powerful workstation and server-grade graphics cards to date.
Available in both desktop and server editions, this new GPU family delivers performance for professionals pushing the boundaries of creativity, engineering, and data science.
NVIDIA RTX PRO™ 6000 Blackwell Technical Specifications
| |
NVIDIA RTX PRO 6000 Blackwell Server Edition |
NVIDIA RTX PRO 6000 Blackwell Workstation Edition | NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition |
| GPU Memory | 96 GB GDDR7 with error-correcting code (ECC) | 96 GB GDDR7 with error-correcting code (ECC) | 96 GB GDDR7 with error-correcting code (ECC) |
| Display Ports | 4x DisplayPort 2.1 | 4x DisplayPort 2.1 | 4x DisplayPort 2.1 |
| Power Consumption | 400-600 W | 600 W | 300 W |
| Graphics Bus | PCIe Gen 5 x16 | PCIe Gen 5 x16 | PCIe Gen 5 x16 |
| Form Factor | 4.4” (H) x 10.5” (L) dual slot | 5.4” (H) x 12” (L) dual slot | 4.4” (H) x 10.5” (L) dual slot |
| Thermal | Passive | Double flow-through | Active |
RTX PRO™ 6000 Blackwell vs. Ada Generation?
The RTX 6000 Blackwell represents a significant leap forward from the Ada Lovelace RTX 6000 series, bringing key architectural, performance, and AI-focused improvements that make it a powerhouse for next-gen workloads.
Performance
With next-gen Tensor Cores, the RTX 6000 Blackwell offers faster FP8, BF16, and INT8 performance, which is crucial for modern AI workflows like LLMs (Large Language Models), diffusion models, and physics-informed ML.
Bandwidth
The RTX 6000 Blackwell is equipped with 96 GB of ultra-fast GDDR7 memory, doubling the capacity of the 48 GB GDDR6 ECC found in the Ada version.
This upgrade provides significantly higher memory bandwidth and enables seamless performance on large-scale AI models, 3D scenes, and complex simulations.
Power Efficiency
The RTX 6000 Blackwell has a TDP of 600W max, doubling the 300W of the RTX 6000 Ada generation.
This increase supports its significantly higher performance, including more cores and faster memory.
For professionals needing a more power-efficient solution, the RTX 6000 Blackwell Max-Q offers similar performance at a reduced 300W TDP.
Scalability
Thanks to NVLink Gen 5, the Blackwell GPUs enable even faster data sharing between multiple GPUs, critical for workloads requiring large model parallelism.
This wasn’t available on all Ada workstation implementations.
RTX PRO™ 6000 Series GPU Use Cases
The RTX PRO™ 6000 Series is engineered for professionals working in the most demanding fields. From AI development to scientific research, its capabilities unlock new levels of performance.
Ideal Use Cases:
- Deep Learning and AI: the RTX 6000 GPU for deep learning is optimized for FP8 workloads, enabling faster model training and inference.
- 3D Rendering and VFX: real-time ray tracing and enhanced CUDA performance make it perfect for animators and rendering farms.
- Data Science and Simulation: high memory bandwidth and compute power allow seamless analysis of large datasets.
- Engineering and CAD: smooth visualization of complex models with ultra-high polygon counts.
Interested? Get in touch with our sales team! We have servers and configurations available now.
2CRSi is NVIDIA Elite Partner.
