SKIP TO MAIN CONTENT
01HPE

P9U90A NVIDIA Tesla M4 4GB computational accelerator

AI Infrastructure

The HPE P9U90A is an NVIDIA Tesla M4 4 GB accelerator. Its memory bandwidth is 88 GB/s. It is rated for 2.2 TFLOPS of FP32 compute. Request a quote and an engineer checks it against your build.

Also known as · HPE NVIDIA Tesla M4 4GB

Price
Request quote
REQUEST QUOTE469.476.5010
NO CARD · NO COMMITMENT · NET-30 / GSA / PO ALL ACCEPTED
1-YR ZAYNTEK WARRANTY
02PRODUCT OVERVIEW

Overview

HPE P9U90A is an NVIDIA Tesla M4 4 GB accelerator.

Its memory bandwidth is 88 GB/s. It is rated for 2.2 TFLOPS of FP32 compute. It is a low-profile 50–75 W card for hyperscale inference.

Key facts

  • GPU: NVIDIA Tesla M4
  • Memory: 4 GB
  • Part description: NVIDIA Tesla M4 4GB computational accelerator

Detailed specifications

  • FP64: 0.07 TFLOPS
  • FP32: 2.2 TFLOPS
  • GPU memory: 4 GB GDDR5
  • Memory bandwidth: 88 GB/s
  • Max power: 50–75 W
  • Host interface: PCIe Gen3
  • Form factor: Low-profile PCIe
  • Cooling: Passive

Add it to a quote with your server model and an engineer checks compatibility before you order.

03TECHNICAL DATA

Specifications

Compute

HPE P9U90A Compute
FP640.07 TFLOPS
FP322.2 TFLOPS

Memory

HPE P9U90A Memory
GPU memory4 GB GDDR5
Memory bandwidth88 GB/s

Power & form factor

HPE P9U90A Power & form factor
Max power50–75 W
Host interfacePCIe Gen3
CoolingPassive
HPE P9U90A specifications
CategoryAI Infrastructure
Warranty1-year ZaynTek

§ 04Related — AI Infrastructure

View all

NOTE · Condition is shown on each listing where an offer states it; ask for it on any quote. Units marked REFURBISHED are fully tested and carry a 1-year ZaynTek warranty.