Brand: NVIDIA | Category: GPUs
SKU: NVIDIA-900-5G133-2500-000 | Part #: 900-5G133-2500-000 | MPN: 900-5G133-2500-000
Contact for Pricing — Request a Quote
PCIe 5.0 x192 connectivity and 575 W maximum power consumption are the defining specifications for integrating the NVIDIA RTX PRO 6000 Blackwell Server Edition into enterprise-grade data center infrastructure. This server-class GPU (part number 900-5G133-2500-000) delivers 48 GB of GDDR7 memory paired with 960 GB/s memory bandwidth, enabling sustained throughput for large-scale compute jobs. Built on NVIDIA's Blackwell architecture and housed in a PCIe 5.0 full-height double-width form factor, the RTX PRO 6000 achieves peak tensor performance of 1,457 TFLOPS at TF32 precision and 5,832 TFLOPS at FP8—positioning it as a high-performance accelerator for demanding workloads. Power delivery via dual 6-pin and 16-pin server-class connectors ensures compatibility with standard enterprise PSU configurations. NVIDIA backs the unit with a 3-year manufacturer warranty, with extended coverage available for critical deployments.
AI infrastructure teams, HPC centers, and machine learning operations teams select the RTX PRO 6000 Blackwell Server Edition for LLM training, inference, and high-performance computing workloads where memory capacity and bandwidth directly impact throughput and model convergence speed. Organizations running production large language model services, scientific computing simulations, and enterprise AI pipelines depend on this GPU to handle both dense compute and memory-intensive tasks without architectural compromise. The combination of 48 GB GDDR7 memory and PCIe 5.0 lanes reduces bottlenecks in multi-GPU clusters, where aggregate bandwidth and per-node capacity drive overall system efficiency.
To procure the NVIDIA RTX PRO 6000 Blackwell Server Edition (900-5G133-2500-000) with full technical validation and support, submit a Request for Quote through Omnixon Global today.
```| Brand | NVIDIA |
| Category | GPUs |
| SKU | NVIDIA-900-5G133-2500-000 |
| Part Number | 900-5G133-2500-000 |
| Condition | New |
| Capacity | 96GB |
| Interface | PCIe Gen5 |
| Power Connector | 12V-2x6 / 16-pin (server-class) |
| Warranty | 3-year manufacturer warranty (extended available) |
| Manufacturer Part Number | 900-5G133-2500-000 |
| Product Line | RTX PRO 6000 Blackwell |
| Form Factor | PCIe 5.0 Full-Height Double-Width GPU |
| GPU Memory | 48 GB GDDR7 |
| Memory Bandwidth | 960 GB/s |
| Peak Tensor Performance (TF32) | 1,457 TFLOPS |
| Peak Tensor Performance (FP8) | 5,832 TFLOPS |
| Interface | PCIe 5.0 x192 |
| Maximum Power Consumption | 575 W |
| Edition | Server Edition |
| Architecture | NVIDIA Blackwell |
| Primary Use Cases | LLM Training, Inference, HPC Workloads |
Reference servers include Dell PowerEdge XE9680 / XE9712, HPE Cray XD670, Lenovo ThinkSystem SR685a / SR675 V3, Supermicro AS-A21GE / SYS-821GE, Gigabyte G593 / G894, ASUS ESC. Share your target platform in the RFQ and we will confirm chassis-to-GPU compatibility and recommended NIC pairing.
Highly model-dependent. L40S / RTX-class: typically 3-6 weeks. H100/H200/B200 in SXM form factor: 12-16 weeks for whole-platform allocations. We quote authorised-channel ETAs only — no grey-market promises.
Yes — Omnixon stocks the full NVIDIA networking lineup (Quantum-2 / Quantum-X InfiniBand, Spectrum-X Ethernet, ConnectX NICs, BlueField DPUs) so we can quote a complete training-cluster BOM, not just the GPUs.
Yes. We hold authorised channels for NVIDIA AI Enterprise software subscriptions. Add it to your RFQ and we quote node-aligned licensing along with the hardware.