Intel Data Center GPU Max 1350 OAM 96GB HBM2e
Brand: Intel | Category: GPUs
SKU: XPGEDCGPU1350X | Part #: XPGEDCGPU1350X | MPN: XPGEDCGPU1350X
Contact for Pricing — Request a Quote
About the Intel Data Center GPU Max 1350 OAM 96GB HBM2e
AI training and inference workloads demand accelerators built on cutting-edge compute architectures, and the Intel Data Center GPU Max 1350 OAM delivers exactly that foundation through its Intel Xe-HPC microarchitecture. This OAM (Open Accelerator Module) form factor GPU pairs 128 Xe Vector Engines with 128 XMX Engines and 128 Ray Tracing Units, enabling parallel processing at scale across diverse workload profiles. With 96 GB of HBM2e memory and a memory bandwidth of 3.2 TB/s, the XPGEDCGPU1350X accelerates both compute-bound and memory-bound operations critical to modern data centers and HPC environments.
Built on Intel 7 process technology for compute tiles and manufactured with EMIB and Foveros interconnect technology, this Intel accelerator integrates multi-tile architecture designed for maximum throughput in production AI and scientific computing scenarios. The GPU connects via PCIe 5.0 x16 and operates within a 450 W thermal design power envelope, making it suitable for dense deployment in enterprise data center racks. Programming flexibility comes through support for oneAPI, SYCL, OpenMP offload, and OpenCL, while Linux operating systems (RHEL, SLES, Ubuntu) ensure seamless integration into existing infrastructure. For IT procurement teams and AI infrastructure engineers evaluating high-performance accelerators, this solution addresses demanding compute requirements across training, inference, and traditional HPC workloads. Contact Omnixon Global to request a quotation for the Intel Data Center GPU Max 1350 OAM (part number XPGEDCGPU1350X) and explore how this accelerator fits your enterprise architecture.
Key Specifications
- Manufacturer: Intel
- Manufacturer Part Number: XPGEDCGPU1350X
- Microarchitecture: Intel Xe-HPC (Ponte Vecchio)
- Memory Capacity: 96 GB HBM2e
- Memory Bandwidth: 3.2 TB/s
- Host Interface: PCIe 5.0 x16
- Thermal Design Power: 450 W
- Operating System Support: Linux (RHEL, SLES, Ubuntu)
Technical Specifications
| Brand | Intel |
| Category | GPUs |
| SKU | XPGEDCGPU1350X |
| Part Number | XPGEDCGPU1350X |
| Condition | New |
| Manufacturer Part Number | XPGEDCGPU1350X |
| Product Name | Intel Data Center GPU Max 1350 OAM |
| Microarchitecture | Intel Xe-HPC (Ponte Vecchio) |
| Form Factor | OAM (Open Accelerator Module) |
| Memory Capacity | 96 GB HBM2e |
| Memory Type | HBM2e (High Bandwidth Memory 2e) |
| Memory Bandwidth | 3.2 TB/s |
| Xe-HPC Stacks | 1 |
| Xe Vector Engines | 128 |
| Xe Matrix Extensions (XMX) Engines | 128 |
| Ray Tracing Units | 128 |
| Thermal Design Power (TDP) | 450 W |
| Process Node | Intel 7 (compute tiles), TSMC (other tiles), multi-tile via EMIB/Foveros |
| Interconnect Technology | EMIB (Embedded Multi-die Interconnect Bridge) and Foveros |
| Host Interface | PCIe 5.0 x16 |
| Programming Models Supported | oneAPI, SYCL, OpenMP offload, OpenCL |
| Operating System Support | Linux (RHEL, SLES, Ubuntu) |
| Target Segment | Data Center, HPC, AI Training and Inference |
| Compliance | OAM Compliance Specification |
Frequently Asked Questions about Intel Data Center GPU Max 1350 OAM 96GB HBM2e
What server platforms accept the Intel Data Center GPU Max 1350 OAM 96GB HBM2e?
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.
How long is the lead time on AI GPUs?
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 genuine-channel ETAs only — no grey-market promises.
Do you supply matched networking (Quantum InfiniBand / Spectrum-X)?
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.
Can you help with NVIDIA AI Enterprise licensing?
Yes. We hold genuine-channels for NVIDIA AI Enterprise software subscriptions. Add it to your RFQ and we quote node-aligned licensing along with the hardware.