Made in the U.S.A
AMD Instinct MI350P
BUILT FOR THE INFRASTRUCTURE YOU HAVE. READY FOR THE AI YOU NEED
WHY BUY AMD SOLUTIONS FROM KOI COMPUTERS?
As a 2026 AMD Partner Network Gold partner, Koi Computers has deployed hundreds of AMD EPYC™ GPU server and HPC solutions to the Federal Government accelerating AI, Deep Learning, and other critical missions.
- Validated and optimized compute, storage, and rack-level solutions that are TAA & FIPS compliant;
- System integration, on-site installation, and start-up deployment services available;
- ISO 9001:2015 certified and 3rd party audited for quality assurance;
- ISO/SEC 20243:2018 (Open Group) Certification for Product Integrity & Supply Chain Security
- Satisfactory or above performance ratings in CPARS – no adverse performance ever;
- Successfully deployed 5 Supercomputers named to the Top500
UPGRADE YOUR INFRASTRUCTURE, NOT YOUR DATA CENTER
Designed for the PCIe® form factor in a full-height, full-length, dual-slot configuration, the AMD Instinct MI350P PCIe® card drops into mainstream air-cooled servers—no specialized cooling, no rack redesigns, no building from scratch. Just the Gen AI and agentic AI performance business demands, deployed in current data centers..
ENTERPRISE AI, OPEN BY DESIGN
Adopting the AMD Instinct MI350P shouldn’t mean rebuilding AI software stacks from scratch. The accelerator pairs with a fully integrated, open, standards-based AI platform designed to make migration seamless—facilitating moves from current GPU infrastructure to the AMD Instinct MI350P without code rewrites, retraining, or rearchitecting deployments.
LEADERSHIP TCO FOR ENTERPRISE AI
Hyperscalers optimize for raw scale. Enterprises optimize for value. The AMD Instinct MI350P PCIe® card is built for the latter—delivering leadership total cost of ownership through every dimension of
AI deployment.
AI deployment.
AMD ENTERPRISE AI SOFTWARE
In combination with the AMD GPU Operator, the AMD Enterprise AI reference stack offers a reference architecture for how to build your own custom development stack integrating preferred ecosystem partner software to accelerate the deployment of enterprise AI solutions on AMD based hardware.
AMD INSTINCT™ MI350P SERIES GPUs
The AMD Instinct™ MI350P PCIe® card brings generative and agentic AI workloads into existing data centers. Built on a standard PCIe form factor, the AMD Instinct MI350P integrates seamlessly into existing infrastructure without the cost or complexity of purpose-built AI systems. The result is a straightforward path to exceptional AI performance, cost savings, and an enterprise-ready software stack designed to reduce time to value.
G494-ZB0-AAP2
4U
Supports
Dual Socket AMD EPYC™ 9005 processors
24 x DIMM slots (DDR5-6400)
5 x1 Redundant 80 PLUS 3000W Titanium PSU
8 x NVMe + 2 x NVMe /SATA/SAS
8 x PCIe Gen5 x16 (LP, HL) GPU group
2 x PCIe Gen5 x16 (FH, HL) - CPU 1 & 2 extension; 1x PCIe Gen5 x8 (FH, HL) - CPU 1 directly
G294-Z42-AAP2
2U
Dual AMD EPYC™ 9005/9004 server processors
24 x DIMM slots (up to DDR5-6400)
Dual 3000 W 80 PLUS Titanium redundant power supplies
4 x 2.5" Gen5 NVMe/SATA/SAS-4 hot-swap bays
2 x LP PCIe x16 slots with Gen5 x16 lanes for add-in cards
8 x FHFL PCIe x16 slots with Gen5 x16 lanes for GPUs
G494-ZB4-AAP2
4U
Dual Socket AMD EPYC™ 9005/9004 series processors
48 x DIMM slots (DDR5-5200)
4 x 3000 W 80 PLUS Titanium redundant power supply
2 x M.2 PCIe Gen3 x4
2 x FHHL PCIe Gen5 x16
8 x FHHL PCIe Gen5 x16 for GPUs
XV23-ZX0-AAJ1
2U
Dual AMD EPYC™ 9005/9004 server processors
24 x DIMM slots (DDR5-6400)
3+1 2000 W 80 PLUS Titanium redundant power supplies
3 x M.2 slots with PCIe Gen3 x2 and x1
3 x FHFL PCIe x16 slots with Gen5 x16 lanes
4 x FHFL PCIe x16 slots with Gen5 x16 lanes for GPUs
G294-Z22-AAP2
2U
Dual AMD EPYC™ 9005/9004 server processors
12 x DIMM slots (DDR5-6400)
Dual 3000 W 80 PLUS Titanium redundant power supplies
2 x M.2 slots with PCIe Gen5 x4
2 x LP PCIe x16 slots with Gen5 x16 lanes for add-in cards
8 x FHFL PCIe x16 slots with Gen5 x16 lanes for GPUs