inf2.48xlarge by Amazon Web Services

inf2.48xlarge is a AWS Inferentia Gen2 48xlarge server offered by Amazon Web Services with 192 vCPUs, 768 GiB of memory and 0 GB of storage. The pricing starts at 1.2981 USD per hour.
192 vCPU
768 GiB Memory
Spare SCore
286,883
(All-cores)
2,992
(Single-core)

Specifications

Server Metadata

Vendor ID
aws
Name
inf2.48xlarge
Description
AWS Inferentia Gen2 48xlarge
Family
inf2
Hw Virt
Status
active
Observed At
2026-10-30T01:01:11.048499

Availability

REGION / IDSPOTONDEMAND
Ohio (US) / us-east-2
1.573967 USD/h 12.9813 USD/h
Northern Virgina (US) / us-east-1
3.377175 USD/h 12.9813 USD/h
Oregon (US) / us-west-2
2.3839 USD/h 12.9813 USD/h
Stockholm (SE) / eu-north-1
1.69015 USD/h 14.2794 USD/h
Dublin (IE) / eu-west-1
3.662333 USD/h 16.2266 USD/h
Mumbai (IN) / ap-south-1
4.4728 USD/h 16.8757 USD/h
Sydney (AU) / ap-southeast-2
12.62545 USD/h 16.8757 USD/h
Paris (FR) / eu-west-3
2.2216 USD/h 18.1738 USD/h
Singapore (SG) / ap-southeast-1
5.46995 USD/h 18.1738 USD/h
London (GB) / eu-west-2
4.7869 USD/h 19.4719 USD/h
Tokyo (JP) / ap-northeast-1
19.4719 USD/h 19.4719 USD/h
Frankfurt (DE) / eu-central-1
19.4719 USD/h 19.4719 USD/h
Sao Paulo (BR) / sa-east-1
2.2068 USD/h 22.0681 USD/h

Processor

vCPUs
192
Hypervisor
nitro
CPU Allocation
Dedicated
CPU Cores
96
CPU Speed
3.6 GHz
CPU Architecture
x86_64
CPU Manufacturer
AMD
CPU Family
EPYC
CPU Model
7R13
CPU L1D Cache
32 KiB
CPU L1D Cache Total
3 MiB
CPU L1I Cache
32 KiB
CPU L1I Cache Total
3 MiB
CPU L2 Cache
512 KiB
CPU L2 Cache Total
48 MiB
CPU L3 Cache
32 MiB
CPU L3 Cache Total
384 MiB
CPU Flags
fpu, vme, de, pse, tsc, msr, pae, mce, cx8, apic, sep, mtrr, pge, mca, cmov, pat, pse36, clflush, mmx, fxsr, sse, sse2, ht, syscall, nx, mmxext, fxsr_opt, pdpe1gb, rdtscp, lm, constant_tsc, rep_good, nopl, nonstop_tsc, cpuid, extd_apicid, aperfmperf, tsc_known_freq, pni, pclmulqdq, monitor, ssse3, fma, cx16, pcid, sse4_1, sse4_2, x2apic, movbe, popcnt, aes, xsave, avx, f16c, rdrand, hypervisor, lahf_lm, cmp_legacy, cr8_legacy, abm, sse4a, misalignsse, 3dnowprefetch, topoext, perfctr_core, ssbd, ibrs, ibpb, stibp, vmmcall, fsgsbase, bmi1, avx2, smep, bmi2, invpcid, rdseed, adx, smap, clflushopt, clwb, sha_ni, xsaveopt, xsavec, xgetbv1, clzero, xsaveerptr, rdpru, wbnoinvd, arat, npt, nrip_save, vaes, vpclmulqdq, rdpid
Ecpus
95.9
Scalability
49.95

System Resources and Accelerators

MEMORY
Memory Amount
768 GiB
Memory Amount Actual
768 GiB
Memory Generation
DDR4
Memory Speed
3200 Mhz
ACCELERATORS
Accelerator Count
0
Accelerator Memory Min
0 MiB
Accelerator Memory Total
0 MiB
Accelerators
    STORAGE
    Storage Size
    0 GB
    Storages
      NETWORK
      Network Speed Baseline
      100 Gbps
      Network Speed Max
      100 Gbps
      Network Storage Speed Baseline
      60 Gbps
      Network Storage Speed Max
      60 Gbps
      Inbound Traffic
      0 GB/month
      Outbound Traffic
      0 GB/month
      IPv4
      0

      Server Description

      A high-density compute instance featuring 192 vCPUs, 100 Gbps networking, and top-tier multi-core performance for demanding parallel workloads.

      General PurposeCompute Optimized

      Amazon Web Services inf2.48xlarge is an Inferentia Gen2 instance featuring 192 vCPUs and 96 physical cores powered by an AMD EPYC 7R13 processor running at 3.6 GHz. Built on the AWS Nitro hypervisor, it includes 768.0 GB of DDR4 memory and a 100 Gbps baseline network interface, with no local storage. The instance delivers top-tier multi-core CPU performance, ranking in the top 10% on stress-ng and PassMark CPU tests, alongside strong cached memory bandwidth and high-throughput Redis SET operations. However, its PassMark memory score is in the bottom 10%. The inf2.48xlarge is highly effective for multi-threaded compute tasks, software compilation, ray tracing, and high-bandwidth network services, though storage-dependent workloads must rely on network-attached volumes.

      Economics

      Average Price per Region

      Prices per Zone

      Lowest Prices

      Performance

      Workload Profiles

      4.4Score
      Precomputed compound score for Cache Intensive workloads. A weighted average (geometric mean) of benchmark scores compared to their medians: score = ∏ (x_i / m_i)^(w_i / Σw). The score of 1.0 represents a synthetic baseline server with the median performance of each component benchmark; 0.5 means roughly half the performance; and 2.0 means twice the performance of that reference profile. Component weights: 50% Redis RPS (pipeline=1, SET), 20% Redis RPS (pipeline=16, SET), 10% PassMark Memory Mark (composite), 10% Memory bandwidth (read, 16 MB ~ L3), 10% PassMark single-thread CPU. Rationale for component selection: In-memory key-value store workload, mixing direct Redis performance metrics with memory speed and latency benchmarks, and single-core CPU performance profiles.
      Component Score Weight Impact
      Raw Reference Normalized Target Actual
      Redis RPS (pipeline=1, SET) 15,350,843 ops/sec 1,919,289 ops/sec 7.9 50.00% 50.00% +180.00%
      Redis RPS (pipeline=16, SET) 106,276,782 ops/sec 11,728,058 ops/sec 9.2 20.00% 20.00% +56.00%
      PassMark Memory Mark (composite) 1,174 2,416 0.5 10.00% 10.00% -6.70%
      Memory bandwidth (read, 16 MB ~ L3) 273,867 MB/sec 108,466 MB/sec 2.5 10.00% 10.00% +9.60%
      PassMark single-thread CPU 1,980 Mops/s 2,365 Mops/s 0.83 10.00% 10.00% -1.80%

      Memory Bandwidth

      Compression

      OpenSSL

      Geekbench Single-Core

      Score: 1,677

      Geekbench Multi-Core

      Score: 15,331

      Passmark CPU Scores

      BENCHMARKSCORE
      Mark
      		 95,525
      Compression
      		 2,982,357
      Encryption
      		 221,586
      Extended Instructions
      		 153,692
      Floating Point Maths
      		 483,941
      Integer Maths
      		 880,361
      Physics
      		 14,535
      Prime Numbers
      		 1,050
      Single Threaded
      		 1,980
      String Sorting
      		 367,059

      Passmark Memory Scores

      BENCHMARKSCORE
      Memory Mark
      		 1,174
      Database Operations
      		 16,815
      Memory Latency
      		 73
      Memory Read Cached
      		 10,071
      Memory Read Uncached
      		 3,053
      Memory Write
      		 2,914

      Stress-ng Raw Scores

      Stress-ng Relative Multicore Performance

      LLM Inference Speed for Prompt Processing

      LLM Inference Speed for Text Generation

      Static Web Server

      Redis

      Alternatives

      Servers of the Same Family

      INSTANCEvCPUsMEMORYAccelerators
      inf2.xlarge 416 GiB0
      inf2.8xlarge 32128 GiB0
      inf2.24xlarge 96384 GiB0

      Similar Servers

      INSTANCEVENDORvCPUsMEMORYAccelerators

      inf2.48xlarge FAQs