1 1

Cisco

Cisco UCSX-GPU-RTXP6000 | UCS X-Series RTX Pro 6000 96GB GPU, 600W, 2-Slot FHFL

UCSX-GPU-RTXP6000

Stock Status: Enquire

Request Quote
Sale Sold out
Shipping calculated at checkout.
20-min response

Need a quote or bulk pricing? Answered within 20 minutes, Sydney business hours.

Description

The UCS X-Series RTX Pro 6000 is an NVIDIA RTX PRO 6000 Blackwell Server Edition GPU for the Cisco UCS X580p PCIe Node. It is built for agentic and physical AI, generative AI inference and fine-tuning, scientific computing and simulation, medical imaging, 3D rendering and content creation, VDI with vGPU, and edge-to-core hybrid AI deployments.

The card is a passive, 2-slot full-height full-length PCIe 5.0 x16 GPU with 96GB of GDDR7 ECC memory on a 512-bit interface, delivering up to 1597 GB/s of memory bandwidth. It has 24,064 CUDA cores, 752 5th-generation Tensor Cores and 188 4th-generation RT Cores, with peak figures of 120 TFLOPS FP32 and 4 PFLOPS FP4 AI performance. Board power is 600W and can be power-capped to 450W. Up to four GPUs fit in a single X580p node, and GPUs in one assignment must be identical in type.

This PID is the UCS X-Series version, used with UCS X210c M8, X215c M8 and X210c M7 compute nodes in a UCS X9508 chassis with UCS X9516 X-Fabric modules. An NVIDIA software licence selection is required when the GPU is ordered on the X580p, with quantity equal to the number of GPUs, or you can opt out. The GPU is managed through Cisco Intersight in Intersight Managed Mode.

Enquire with XS Network Tech for new or used pricing and availability.

Features

- Blackwell architecture GPU for agentic and physical AI, generative AI inference and fine-tuning
- Advanced scientific computing, simulation and rendering
- High-fidelity 3D graphics, video and content creation
- Hybrid AI and machine learning workflows from edge to core
- Managed through Cisco Intersight with policy-based GPU provisioning
- Shareable across up to two compute nodes from a single X580p via UCS X-Fabric
- Passive cooling that relies on chassis airflow
- Power capping from 600W down to 450W

Warranty

All products sold by XS Network Tech include a 12-month warranty on both new and used items. Our in-house technical team thoroughly tests used hardware prior to sale to ensure enterprise-grade reliability.

All technical data should be verified on the manufacturer data sheets.

View full details

specs-tabs

Technical Specifications

FAQs

Technical Specifications

- GPU: NVIDIA RTX PRO 6000 Blackwell Server Edition
- GPU memory: 96 GB GDDR7 with ECC
- Memory interface: 512-bit
- Memory bandwidth: Up to 1597 GB/s
- CUDA cores: 24,064
- Tensor Cores: 752 (5th generation)
- RT Cores: 188 (4th generation)
- FP32 performance: 120 TFLOPS (peak)
- TF32 Tensor Core performance: 234 TFLOPS (peak)
- FP16 and BF16 Tensor Core performance: 1 PFLOPS (peak)
- FP8 Tensor Core performance: 2 PFLOPS (peak)
- FP4 AI performance: 4 PFLOPS (peak)
- RT Core performance: 355 TFLOPS (peak)
- Host interface: PCIe 5.0 x16
- NVLink support: No
- Power consumption: 600 W (can be power-capped to 450 W)
- Form factor: 2-slot full-height full-length (FHFL), passive cooling
- Maximum per UCS X580p PCIe Node: 4
- Riser slot compatibility: Cage A and B, Slots 1, 2, 3, 4
- Supported compute nodes: UCS X210c M8, UCS X215c M8, UCS X210c M7
- Operating temperature (X580p node): 10 to 27 C
- Operating humidity (X580p node): 5% to 93% noncondensing

FAQs

Q: Does this GPU require a software licence?
A: An NVIDIA software licence selection is required when the GPU is ordered on the X580p, with quantity equal to the number of GPUs. Options include AI Enterprise Essentials, vGPU licences or an opt-out selection. The node also needs a minimum Cisco Intersight Essentials licence per server in IMM mode.

Q: What power does the GPU draw?
A: It has a maximum board power of 600W, which can be power-capped to 450W. Cooling is passive, so it depends on chassis airflow. The Cisco UCS Power Calculator gives configuration-specific figures.

Q: Which Cisco systems does this GPU work with?
A: It is used in the Cisco UCS X580p PCIe Node (UCSX-580P-U) inside a UCS X9508 chassis with UCS X9516 X-Fabric modules. It pairs with UCS X210c M8, X215c M8 and X210c M7 compute nodes, and works with 4th, 5th and 6th generation Fabric Interconnects in Intersight Managed Mode.

Q: How many of these GPUs fit in one node?
A: Up to four per UCS X580p PCIe Node, across Cage A and B, Slots 1 to 4. GPUs in a single assignment must be identical in type.

Q: Does it support NVLink?
A: No. This GPU does not support NVLink bridging.

Q: How does this differ from the C-Series and other platform versions?
A: UCSX-GPU-RTXP6000 is the UCS X-Series PID. The same GPU is sold under different PIDs for other platforms, such as UCSC-GPU-RTXP6000 for C-Series, CAI-GPU-RTXP6000 for C845A and UCSXE-GPU-RTXP6000 for XE150c.

Related Products

View all Graphics Card | GPU

Recently Viewed

  • Request a Quote

    Looking for competitive pricing? Submit a request, and our team will provide a tailored quote that fits your needs.

  • Contact Us Directly

    Have a question or need immediate assistance? Call us for expert advice and real-time support.

    Call us Now  

  • Contact Us Directly

    Have a question or need immediate assistance? Call us for expert advice and real-time support.

    Contact us