Consumercarts
AMD EPYC 7373X processor (Milan-X, Socket SP3)
AMD

AMD EPYC 7373X

View on amd.com
Brand:
AMD
Form Factor:
Server & Enthusiast
CPU Socket:
SP3
Core/Thread Count:
16C/32T
TDP:
240W
CPU Series:
EPYC 7003 Series

Consumercarts is reader-supported. Learn more

Disclaimer: The price and specification of the given product may differ from the actual product. Please confirm on the retailer's website before buying.

Manufacturer: AMD Last updated: October 9, 2026

Report an error

AMD EPYC 7373X Review

AMD EPYC 7373X server processor, 16C/32T 3.05 GHz base clock, boosting up to 3.8 ghz, 240W TDP, SP3 platform.

The AMD EPYC 7373X is an AMD EPYC server processor in the EPYC 7003 Series lineup, offering 16C/32T running at 3.05 GHz base clock, boosting up to 3.8 ghz with 768 MB of L3 cache, within a 240W power envelope on AMD's SP3 platform in 1P / 2P configurations.

Performance and Architecture

With 16C/32T, the AMD EPYC 7373X runs at 3.05 GHz base clock, boosting up to 3.8 ghz. It carries 768 MB of L3 cache to keep large working sets close to the cores. As part of AMD's EPYC server lineup, it targets dense multi-socket compute, virtualization, and datacenter workloads.

Platform and Memory

The AMD EPYC 7373X is built for AMD's SP3 platform, supporting 1P / 2P socket configurations. It supports DDR4 memory across 8 channels (Up to 3200 MT/s). I/O runs over PCIe 4.0 x128.

Comparison to Alternatives

Within the EPYC 7003 Series lineup in this catalog, the AMD EPYC 7373X is positioned by its core count, clock speed, and TDP relative to other EPYC parts in the same generation. Buyers should weigh core density against per-core clock speed depending on whether the target workload favors many parallel threads or fewer, faster ones.

Our verdict

Reviewed by Burooj Alam. Rated by us, not by users.

Best for

  • Datacenter and cloud compute
  • Virtualization-dense server workloads
  • Multi-socket enterprise deployments

Not for

  • Consumer desktop builds
  • Single-threaded latency-sensitive workloads on lower core-count SKUs
  • Budget-constrained small business servers

Specifications

What these specs mean

Identity

Release Year2021

Cores and threads

Core Count16 cores
Thread Count32 threads

Clocks

Base Clock3.05 GHz
Boost ClockUp to 3.8 GHz

Cache

L3 Cache768 MB

Platform

SocketSP3
Multi-Socket Support1P / 2P

Memory

Memory Channels8 channels
Max Memory Speed3200 MT/s
Memory Bandwidth204.8 GB/s

PCIe and IO

PCIe Lanes128 lanes
PCIe GenerationPCIe 4.0

Power and thermals

TDP240 W
Configurable TDP Max225-280W

SPEC CPU 2017

Best published result on spec.org as of 23 Sept 2026.

1-chip system
SPECrate® 2017 Integer (base)179

Cores 16 · Tested in ASUS RS520A-E11(KMPA-U16) Server System 3.05 GHz, AMD EPYC 7373X · Published by ASUSTeK Computer Inc., Jul-2022

Full disclosure on spec.org
SPECrate® 2017 Floating Point (base)203

Cores 16 · Tested in ASUS RS520A-E11(KMPA-U16) Server System 3.05 GHz, AMD EPYC 7373X · Published by ASUSTeK Computer Inc., Jul-2022

Full disclosure on spec.org
2-chip system
SPECrate® 2017 Integer (base)357

Cores 32 · Tested in Inspur NF5280A6 (AMD EPYC 7373X) · Published by Inspur Corporation, Sep-2022

Full disclosure on spec.org
SPECrate® 2017 Floating Point (base)401

Cores 32 · Tested in Cisco UCS C245 M6 (AMD EPYC 7373X) · Published by Cisco Systems, May-2022

Full disclosure on spec.org

SPEC®, SPECrate® and SPECspeed® are registered trademarks of the Standard Performance Evaluation Corporation. Copied by permission of the Standard Performance Evaluation Corporation (SPEC). Results retrieved from www.spec.org on 23 Sept 2026.

FAQs

What are the key specifications of the AMD EPYC 7373X?

The AMD EPYC 7373X has 16 cores and 32 threads, a 3.05 GHz base and 3.8 GHz boost clock, 768 MB of L3 cache and a 240 W TDP.

Who is the AMD EPYC 7373X best suited to?

Best suited to: Datacenter and cloud compute; Virtualization-dense server workloads; Multi-socket enterprise deployments.

What are the drawbacks of the AMD EPYC 7373X?

Not the best fit for: Consumer desktop builds; Single-threaded latency-sensitive workloads on lower core-count SKUs; Budget-constrained small business servers.

Compare and read more