Moore Threads · Data center · Accelerator card

Moore Threads MTT S5000 80GB

Exact MTT S5000 reference specifications and AI deployment guidance, separated from FusionPointAI commercial inventory.

VRAM80 GB

Memory technologyHBM2e

Form factorFull height · Full length · Dual slot · PCIe · Accelerator

InterfacePCIe Gen5 x16

PowerNot listed

Current Reference only

Compare GPU

Product images

Moore Threads MTT S5000 80GB — Angled view — Moore Threads MTT S5000 80GB · Angled view MTT S5000 OAM Compute Module (air-cooled)

Specifications

Field	Value
Manufacturer	Moore Threads
Exact SKU	MTT S5000
Product family	Moore Threads MTT S5000
Architecture	PingHu PH100
Generation	MTT S-Series Data Center
Entity type	Accelerator card
Lifecycle	Current
Launch date	Not listed
VRAM	80 GB
Memory technology	HBM2e
Memory bandwidth	1,600 GB/s
Power	Not listed
Interface	PCIe Gen5 x16
Form factor	Full height · Full length · Dual slot · PCIe · Accelerator
Interconnect	MTLink
MIG / partitioning	Not supported
Supported numeric formats	FP64, FP32, TF32, BF16, FP16, FP8, INT8
Software stack	musa, pytorch, vllm, sglang

AI workload suitability

LLM inference

High memory capacity supports large checkpoints and multi-GPU inference when runtime and topology are certified.

Fine-tuning and training

Designed for sustained accelerator workloads; training suitability depends on interconnect, precision and cluster topology.

Image, video and embeddings

FP16 acceleration supports common generative and vector workloads within the listed VRAM envelope.

Power and cooling

The manufacturer does not publish a directly comparable board-power value for this entity.

Multi-GPU use

MTLink is the recorded interconnect path; topology must still be validated at node level.

Software ecosystem

musa, pytorch, vllm, sglang

Product timeline