NVIDIA RTX PRO 6000 Blackwell Server Edition on Nebius AI Cloud
Run AI inference, scientific simulation, and physical AI workloads on NVIDIA’s universal data center GPU with RTX technology, available on Nebius AI Cloud in flexible self-service configurations.
What makes RTX PRO 6000 on Nebius different
GPU performance, optimized in-house
Our GPU clusters are optimized across every layer of the stack — from in-house designed servers to a software layer validated against NVIDIA benchmarks — so your AI workloads run at the highest achievable performance.
AI without operational overhead
We take care of the infrastructure, at any scale. Run containerized experiments with Serverless Jobs or deploy multi-GPU inference clusters, without touching drivers, networking, or cluster configuration.
Security and reliability by design
Every Nebius cluster comes with auto-healing that detects and recovers from hardware failures with minimum possible interruption. Our platform is also built on industry security and compliance standards, so your data and workloads always stay secure.
What teams run on NVIDIA RTX PRO 6000
Cost-effective AI inference
Deploy and serve large language models on a single GPU. With 96 GB of memory, RTX PRO 6000 handles 70B-class models in quantized precision or 30B–40B models at full precision. Up to 4 MIG instances per card allow multiple isolated workloads to share one GPU efficiently.
Visual computing and physical AI
Accelerate workloads at the intersection of AI and graphics: vision-language-action models, embodied reasoning, digital twins, synthetic data generation, and NVIDIA Omniverse applications. Fourth-generation RT Cores and fifth-generation Tensor Cores handle both the AI and rendering sides of these workloads in a single GPU.
Scientific simulation
Run scientific and traditionally CPU-based HPC workloads at FP32 precision, including molecular dynamics, molecular docking, drug discovery, and physics-based simulation. RTX PRO 6000’s compute profile and large memory capacity make it a cost-effective choice for these workloads.
NVIDIA RTX PRO 6000 Blackwell Server Edition specifications
Specification
RTX PRO 6000 Blackwell Server Edition
CUDA parallel processing cores
24,064
NVIDIA Tensor Cores
752 (fifth-generation)
NVIDIA RT Cores
188 (fourth-generation)
Single-precision performance (FP32)
120 TFLOPS
Peak FP4 AI performance
4 PFLOPS
RT Core performance
355 TFLOPS
GPU memory
96 GB GDDR7 with ECC
Memory interface
512-bit
Memory bandwidth
1,597 GB/s
Multi-Instance GPU (MIG)
Up to 4× 24 GB
NVENC / NVDEC / JPEG
4× / 4× / 4×
Confidential Computing
Supported
Secure boot with root of trust
Yes
Graphics bus
PCI Express 5.0×16
Form factor
4.4″ (H) × 10.5″ (L), dual-slot
Thermal solution
Passive; air- and liquid-cooled
Power consumption
Up to 600 W (configurable)
Power connector
1× PCIe CEM5 16-pin
Source: NVIDIA RTX PRO 6000 Blackwell Server Edition official datasheet.
Built for inference, simulation, and physical AI
The RTX PRO 6000 Blackwell Server Edition pairs strong single-precision performance with fourth-generation RT Cores for real-time photorealistic rendering and visualization. This combination makes it one of the few GPU platforms that handles AI inference and graphics-intensive workloads with equal efficiency — from agentic AI and LLM serving to digital twins, synthetic data generation, and Omniverse-based physical AI applications.
With 96 GB of memory and MIG support, it sustains these workloads comfortably: large models fit without quantization trade-offs, and multiple isolated workloads can share a single GPU efficiently.
NVIDIA Exemplar Cloud Validation
NVIDIA Exemplar Cloud Validation
Nebius is a Reference Platform NVIDIA Cloud Partner and holds NVIDIA Exemplar Cloud validation across multiple GPU generations — from NVIDIA H200 to NVIDIA GB300 NVL72. Exemplar Cloud is awarded to providers that demonstrate real-world training performance against NVIDIA’s benchmarking standards, not just peak specifications.
Frequently Asked Questions
The NVIDIA RTX PRO 6000 Blackwell Server Edition is a professional data center GPU built on NVIDIA Blackwell architecture. It combines 96 GB of GDDR7 memory, fifth-generation Tensor Cores with FP4 support, and fourth-generation RT Cores in a passively cooled, dual-slot form factor designed for 24/7 data center operation.
Get started with NVIDIA RTX PRO 6000 Blackwell Server Edition on Nebius
Launch your first instance in minutes or talk to our team to find the right setup for your workload.