AI
AMDDatacenter GPUs
AMD Instinct MI300X
192 GB HBM3 accelerator for LLM training and large-context inference.
MSRP $27,000
Details 
AILimited
Authorized partnerDatacenter GPUsNVIDIA H100 80GB PCIe
SKU H100-PCIE-80GB
Hopper-architecture accelerator for LLM training and inference.
MSRP indication
$30,000
Quote pricing may differ — request below.
Overview
80 GB HBM3 memory at 2 TB/s. Designed for transformer engine workloads and FP8 inference.
Specifications
- Architecture
- Hopper
- Memory
- 80 GB HBM3 @ ~2 TB/s
- Form factor
- Dual-slot PCIe Gen5
- TDP
- 350 W
- FP8 Tensor
- Up to 3,026 TFLOPS
Need it bundled with deployment? Mention in your RFQ — we handle install and support too.
More in Datacenter GPUs
AINew
NVIDIADatacenter GPUs
NVIDIA H200 141GB
Hopper refresh with 141 GB HBM3e — the largest memory in the H-class.
MSRP $32,000
Details NVIDIADatacenter GPUs
NVIDIA L40S 48GB
Universal GPU for AI inference, training, and graphics workloads.
MSRP $8,190
Details