Brand: AMD | Category: GPUs
SKU: 100-300000120 | Part #: 100-300000120 | MPN: 100-300000120
Contact for Pricing — Request a Quote
The AMD Instinct MI350X GPU Accelerator (part number 100-300000120) is a passive OAM form-factor accelerator designed for high-performance data center deployments. Built on the advanced CDNA 4 architecture, this AMD GPU delivers exceptional compute density for AI training, AI inference, and HPC workloads without requiring active blower cooling. The MI350X represents AMD's next-generation data center GPU solution, succeeding the MI300-series with enhanced capabilities and efficiency.
Engineered for enterprise AI infrastructure teams and HPC practitioners, the MI350X combines HBM3E memory technology with AMD Infinity Fabric interconnect architecture to enable seamless multi-GPU scaling and data movement. Support for multiple precision formats—FP64, FP32, TF32, BF16, FP16, and FP8—allows flexible algorithm optimization across diverse machine learning and scientific computing tasks. The accelerator integrates seamlessly with the AMD ROCm open-source software platform and supports industry-standard frameworks including PyTorch, TensorFlow, JAX, OpenMP, and MPI, enabling rapid deployment across heterogeneous computing environments. Available throughout UAE, GCC, EMEA, and APAC regions, the MI350X is positioned as a critical building block for next-generation AI and HPC infrastructure. Contact Omnixon Global to discuss your data center acceleration requirements and request a quotation for the AMD Instinct MI350X GPU Accelerator 100-300000120.
| Brand | AMD |
| Category | GPUs |
| SKU | 100-300000120 |
| Part Number | 100-300000120 |
| Condition | New |
| Manufacturer Part Number | 100-300000120 |
| Product Family | AMD Instinct MI350X |
| GPU Architecture | CDNA 4 |
| Memory Type | HBM3E |
| Form Factor | OAM (OCP Accelerator Module) |
| Supported Precisions | FP64, FP32, TF32, BF16, FP16, FP8 |
| Software Platform | AMD ROCm (open-source) |
| Supported Frameworks | PyTorch, TensorFlow, JAX, OpenMP, MPI |
| Interconnect | AMD Infinity Fabric |
| Target Market | Data center, AI training, AI inference, HPC |
| Product Line | AMD Instinct (Data Center GPU) |
| Generation | MI300-series successor (MI350-series) |
| Regional Availability | UAE, GCC, EMEA, APAC |
The AMD Instinct MI350X GPU Accelerator 100-300000120 accelerates AI/ML training, inference, scientific HPC, and virtualization (vGPU) workloads. Typical deployments include LLM training clusters, computer-vision pipelines, financial risk modeling, and rendering farms.
Key specifications for the AMD Instinct MI350X GPU Accelerator 100-300000120: new condition; gpu support AMD MI300X; ai optimized Yes. Manufacturer part number 100-300000120. For the full datasheet with electrical, environmental, and compliance details, contact our pre-sales engineering team.
The AMD Instinct MI350X GPU Accelerator 100-300000120 requires a PCIe Gen4 or Gen5 x16 slot, server power adequate for the card's TDP, and CUDA/ROCm driver support in your hypervisor or bare-metal OS. Sales engineering will confirm chassis fit (1U/2U/4U), PCIe lane count, and PSU headroom before quoting.