MANGO Alphonso

High EndAMD GPUs

Full Stack Server Optimized for AI

High performance, high efficiency, turnkey AI infrastructure solution.

Product Inquiry

Turnkey Solution. Managed as one.

Full stack AI system. LLMBoost™, BoostX™ RoCE AI and MAC™ tune throughput, cost and operations as one.

Product BriefData Sheet

Full stack AI system.

Built for large scale training and high volume inference, delivered as one validated turnkey stack.

Network and TCO, optimized in house.

LLMBoost™ and onboard BoostX™ RoCE AI offload the data path to raise throughput and lower total cost of ownership.

Unified system. One screen.

MAC™ manages distributed nodes as a single system, with a real time dashboard for easy operations.

Compute and networking, balanced by design.

Compute and networking balanced to keep every GPU fed, never idle.

ALPHONSO High End · single node

System Memory

Gen5 x16

CPU 1

Gen5 x16

2× PCIe Gen5 switch

PCIe Switch

GPU

MI355X

GPU

MI355X

GPU

MI355X

GPU

MI355X

BoostX™

RoCE AI

BoostX™

RoCE AI

BoostX™

RoCE AI

BoostX™

RoCE AI

Up to 400 Gbps per GPU

Ethernet, RoCE v2, UEC-ready

System Memory

Gen5 x16

CPU 2

Gen5 x16

2× PCIe Gen5 switch

PCIe Switch

GPU

MI355X

GPU

MI355X

GPU

MI355X

GPU

MI355X

BoostX™

RoCE AI

BoostX™

RoCE AI

BoostX™

RoCE AI

BoostX™

RoCE AI

Up to 400 Gbps per GPU

Ethernet, RoCE v2, UEC-ready

More performance. Lower cost.

LLMBoost™ lifts throughput on the same GPUs, and BoostX™ RoCE AI runs the cluster on standard Ethernet.

LLMBoost™ vs vLLM on MI355X

Same GPUs. Up to 1.6x more throughput and far lower latency than the latest vLLM.

Deepseek v3.2

OpenOrca, MLPerf inference workload

1.61x

Higher throughput

416x

Lower TTFT latency

Qwen3-Next-80B-A3B

OpenOrca, MLPerf inference workload

1.26x

Higher throughput

51.7x

Lower TTFT latency

Measured on 4x MI355X vs latest vLLM, OpenOrca (MLPerf inference workload). Results on the 8x MI355X node will differ.

UEC-ready features on a standard Ethernet switch.

BoostX™ RoCE AI brings UEC-ready features, packet spraying and programmable congestion control, handling network packets efficiently on the DPU with a standard Ethernet network switch. No premium switch silicon required.

NVIDIA Spectrum-4 SN5400

64 port 400GbE switch

~ $39,650

Arista 7060DX5-64S

64 port 400GbE switch, with BoostX™ RoCE AI

~ $28,000

~29% lower

switch cost per equivalent 400G port, with UEC-ready features moved onto the DPU.

Packet spraying

Traffic spread evenly across links on the DPU, no proprietary switch silicon.

Programmable congestion control

Congestion handled on the DPU and tuned to the workload.

Reuse existing infrastructure

Slots into the Ethernet spine you already run.

Lower operations cost

Standard Ethernet skills, no fabric specialists.

Street pricing referenced from public listings for 64 port 400GbE switches (NVIDIA Spectrum-4 SN5400 and Arista 7060DX5-64S) and varies by vendor, configuration and contract.