l4-90 by OVHcloud

l4-90 is a Cloud GPU (22 vCPUs, 90 GiB RAM, 400 GB NVMe, 1xL4 24 GiB VRAM) server offered by OVHcloud with 22 vCPUs, 90 GiB of memory and 400 GB of storage. The pricing starts at 0.8347 USD per hour.
22 vCPU
90 GiB Memory
400 GB Storage
1 GPU
Spare SCore
68,513
(All-cores)
3,156
(Single-core)

Specifications

Server Metadata

Vendor ID
ovh
Name
l4-90
Description
Cloud GPU (22 vCPUs, 90 GiB RAM, 400 GB NVMe, 1xL4 24 GiB VRAM)
Family
Cloud GPU
Hw Virt
Average Time To Start
152
Status
active
Observed At
2026-10-30T02:01:18.073655

Availability

REGION / IDSPOTONDEMAND
GRA11 (FR) / GRA11
- 0.8347 USD/h

Processor

vCPUs
22
Hypervisor
KVM
CPU Allocation
Dedicated
CPU Cores
22
CPU Speed
2.75 GHz
CPU Architecture
x86_64
CPU Manufacturer
AMD
CPU Family
EPYC
CPU Model
9454
CPU L1D Cache
64 KiB
CPU L1D Cache Total
1 MiB
CPU L1I Cache
64 KiB
CPU L1I Cache Total
1 MiB
CPU L2 Cache
512 KiB
CPU L2 Cache Total
11 MiB
CPU L3 Cache
16 MiB
CPU L3 Cache Total
352 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, syscall, nx, mmxext, fxsr_opt, pdpe1gb, rdtscp, lm, rep_good, nopl, cpuid, extd_apicid, tsc_known_freq, pni, pclmulqdq, ssse3, fma, cx16, pcid, sse4_1, sse4_2, x2apic, movbe, popcnt, tsc_deadline_timer, aes, xsave, avx, f16c, rdrand, hypervisor, lahf_lm, cmp_legacy, svm, cr8_legacy, abm, sse4a, misalignsse, 3dnowprefetch, osvw, perfctr_core, ssbd, ibrs, ibpb, stibp, vmmcall, fsgsbase, tsc_adjust, bmi1, avx2, smep, bmi2, erms, invpcid, avx512f, avx512dq, rdseed, adx, smap, avx512ifma, clflushopt, clwb, avx512cd, sha_ni, avx512bw, avx512vl, xsaveopt, xsavec, xgetbv1, xsaves, avx512_bf16, clzero, xsaveerptr, wbnoinvd, arat, npt, lbrv, nrip_save, tsc_scale, vmcb_clean, pausefilter, pfthreshold, v_vmsave_vmload, vgif, avx512vbmi, umip, pku, ospke, avx512_vbmi2, gfni, vaes, vpclmulqdq, avx512_vnni, avx512_bitalg, avx512_vpopcntdq, la57, rdpid, overflow_recov, succor, fsrm, arch_capabilities
Ecpus
21.7
Scalability
98.64

System Resources and Accelerators

MEMORY
Memory Amount
90 GiB
Memory Amount Actual
88 GiB
ACCELERATORS
Accelerator Count
1
Accelerator Memory Min
24 GiB
Accelerator Memory Total
24 GiB
Accelerator Type
GPU
Accelerator Manufacturer
NVIDIA
Accelerator Family
Ada Lovelace
Accelerator Model
L4
Accelerators
  • NVIDIA Ada Lovelace NVIDIA L4 (Memory amount: 23034, Firmware version: 590.48.01, BIOS version: 95.04.65.00.13, Clock rate: 2040)
STORAGE
Storage Size
400 GB
Storage Type
nvme ssd
Storages
  • 400 GB nvme ssd
NETWORK
Network Speed Baseline
8 Gbps
Network Speed Max
8 Gbps
Inbound Traffic
0 GB/month
Outbound Traffic
0 GB/month
IPv4
1

CPU and System Topology

Server Description

A dedicated AMD EPYC server featuring an NVIDIA L4 GPU for accelerated machine learning inference and compute-intensive workloads.

GPU AcceleratedCompute Optimized

OVHcloud l4-90 is a GPU-accelerated server designed for compute-intensive and machine learning workloads. It features an AMD EPYC 9454 processor with 22 dedicated physical cores, 90.0 GB of RAM, a 400 GB NVMe SSD, and a single NVIDIA L4 GPU with 24 GB of VRAM. Operating on a KVM hypervisor, the server delivers 8 Gbps of baseline network bandwidth. Benchmark data highlights top-tier single-core CPU performance and exceptional LLM inference speeds, particularly for small and medium models. The system also demonstrates strong multi-core capabilities and image processing performance. This hardware profile makes the l4-90 a highly efficient choice for machine learning inference, image processing, and parallel compute tasks.

Economics

Average Price per Region

Prices per Zone

Lowest Prices

Performance

Workload Profiles

1.5Score
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) 2,265,679 ops/sec 1,919,289 ops/sec 1.2 50.00% 50.00% +9.50%
Redis RPS (pipeline=16, SET) 23,327,177 ops/sec 11,728,058 ops/sec 1.9 20.00% 20.00% +14.00%
PassMark Memory Mark (composite) 2,970 2,416 1.2 10.00% 10.00% +1.80%
Memory bandwidth (read, 16 MB ~ L3) 538,809 MB/sec 108,466 MB/sec 4.9 10.00% 10.00% +17.00%
PassMark single-thread CPU 2,971 Mops/s 2,365 Mops/s 1.2 10.00% 10.00% +1.80%

GPU Bandwidth and Latency

Node (all GPUs)
Node baseline directional bandwidth
11.0 GB/s
Node GPU-to-host bandwidth
13.2 GB/s
Node host-to-GPU bandwidth
13.4 GB/s
Single GPU slot
Per-GPU GPU-to-host bandwidth
13.2 GB/s
Per-GPU host-to-GPU bandwidth
13.4 GB/s
Per-GPU host-device latency
645 ns
Benchmark Environment
Driver Version
 535.274.02
CUDA Version (driver)
 12.6
nvbandwidth Version
 v0.5
GPU Count
 1
P2P Supported
No
Efficiency
SM/CE copy efficiency
1.00

Memory Bandwidth

Compression

OpenSSL

Geekbench Single-Core

Score: 2,034

Geekbench Multi-Core

Score: 12,870

Passmark CPU Scores

BENCHMARKSCORE
Mark
 48,609
Compression
 591,932
Encryption
 29,246
Extended Instructions
 54,750
Floating Point Maths
 118,937
Integer Maths
 134,634
Physics
 7,825
Prime Numbers
 447
Single Threaded
 2,971
String Sorting
 85,088

Passmark Memory Scores

BENCHMARKSCORE
Memory Mark
 2,970
Database Operations
 18,913
Memory Latency
 69
Memory Read Cached
 28,330
Memory Read Uncached
 26,507
Memory Write
 23,371

Stress-ng Stressors

CPU integerSingleMultiScaling
atomic
 909.89 421.59 0.46×
bitops
 164,637 3,567,431 21.67×
branch
 218,702 4,774,648 21.83×
funccall
 635.42 13,757 21.65×
funcret
 21,481 467,820 21.78×
hash
 167,890 3,629,236 21.62×
intmath
 68,238 1,471,937 21.57×
nop
 645,076 13,834,233 21.45×
prime
 489,748 10,720,326 21.89×
rotate
 90,771 1,956,987 21.56×
str
 1,063,527 24,435,048 22.98×
wcs
 718,920 15,243,107 21.20×
CryptographySingleMultiScaling
crypt
 3,779 83,365 22.06×
CompressionSingleMultiScaling
jpeg
 902.54 19,428 21.53×
zlib
 71.12 1,443 20.29×
HypervisorSingleMultiScaling
clock
 27,449 236,277 8.61×
clone
 248.63 4,196 16.88×
context
 3,199 69,171 21.62×
cpu-sched
 60.87 1,413 23.21×
eventfd
 39,807 683,300 17.17×
fault
 8,837 137,411 15.55×
fork
 1,616 41,480 25.67×
getrandom
 54,434 1,139,475 20.93×
hrtimers
 278,121 2,149,030 7.73×
mprotect
 599,350 4,184,815 6.98×
mtx
 61,554,154 2,171,228,285 35.27×
munmap
 1,064 33,172 31.19×
null
 366,985 8,087,213 22.04×
pipe
 366,800 7,218,869 19.68×
priv-instr
 389,822 7,741,938 19.86×
resched
 7,620,911 3,783,973 0.50×
switch
 124,615 2,156,935 17.31×
timermix
 125,385 3,155,962 25.17×
tlb-shootdown
 13,595 100,942 7.42×
tsc
 804,058 17,381,062 21.62×
usersyscall
 291,905 6,013,387 20.60×
vdso
 86,240,803 1,868,494,122 21.67×
vfork
 15,115 157,080 10.39×
x86syscall
 4,571,830 100,270,399 21.93×
yield
 12,440,691 13,299,320 1.07×
zero
 2,701,406 59,444,702 22.01×
CPU float/SIMDSingleMultiScaling
eigen
 16,531 365,632 22.12×
expmath
 2,938 61,180 20.83×
fma
 945,192 20,517,399 21.71×
fp
 259.18 5,601 21.61×
fractal
 7.6 165.05 21.72×
hyperbolic
 3,904 84,482 21.64×
logmath
 4,690 101,861 21.72×
matrix
 5,047 110,253 21.85×
matrix-3d
 628.13 11,006 17.52×
monte-carlo
 193.61 4,054 20.94×
mpfr
 1,745 37,976 21.76×
powmath
 1,360 29,528 21.72×
trig
 1,906 41,467 21.75×
veccmp
 19,552 424,169 21.69×
vecfp
 1,864 40,551 21.76×
vecmath
 4,151 89,699 21.61×
vecshuf
 656.34 14,227 21.68×
vecwide
 29,279 643,238 21.97×
AlgorithmsSingleMultiScaling
bitonicsort
 10.8 231.79 21.46×
bsearch
 315.11 6,848 21.73×
bubblesort
 0.8 18 22.50×
fibsearch
 391.7 8,462 21.60×
heapsort
 11.2 242.39 21.64×
hsearch
 11,454 250,631 21.88×
insertionsort
 13.2 286.6 21.71×
malloc
 3,840,502 41,093,893 10.70×
mcontend
 75.7 3,461 45.72×
mergesort
 34.8 760.18 21.84×
qsort
 22.6 492.99 21.81×
skiplist
 7,281 157,579 21.64×
tree
 2 43.97 21.98×
MemorySingleMultiScaling
cache
 447,080 838,103 1.87×
cachehammer
 278,723 5,356,246 19.22×
cacheline
 56.74 1,242 21.89×
l1cache
 445.16 9,669 21.72×
memcpy
 457.48 9,973 21.80×
memrate
 3.71 9.79 2.64×
memthrash
 155.36 824.53 5.31×
ptr-chase
 42,318,574 444,720,322 10.51×
spinmem
 380.98 1,768 4.64×
stream
 274.35 15,980 58.25×
vm
 90,772 294,567 3.25×

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

PostgreSQL heavy read-only throughput

Alternatives

Servers of the Same Family

INSTANCEvCPUsMEMORYAccelerators
rtx5000-28 428 GiB1
t1-le-45 845 GiB1
rtx5000-56 856 GiB2
t2-le-45 1545 GiB1
l40s-90 1590 GiB1
a100-180 15180 GiB1
t1-le-90 1690 GiB2

Similar Servers

INSTANCEVENDORvCPUsMEMORYAccelerators

l4-90 FAQs