g2-standard-4 by Google Cloud Platform

g2-standard-4 is a Graphics Optimized: 1 NVIDIA L4 GPU, 4 vCPUs, 16GB RAM server offered by Google Cloud Platform with 4 vCPUs, 16 GiB of memory and 0 GB of storage. The pricing starts at 0.3068 USD per hour.
4 vCPU
16 GiB Memory
1 GPU
Spare SCore
3,149
(All-cores)
1,198
(Single-core)

Specifications

Server Metadata

Vendor ID
gcp
Server ID
1600004
Name
g2-standard-4
Description
Graphics Optimized: 1 NVIDIA L4 GPU, 4 vCPUs, 16GB RAM
Family
g2
Hw Virt
Average Time To Start
22
Status
active
Observed At
2026-10-26T04:04:46.487057

Availability

REGION / IDSPOTONDEMAND
Ashburn (US) / us-east4
0.3759 USD/h 0.7045 USD/h
Council Bluffs (US) / us-central1
0.423 USD/h 0.7068 USD/h
The Dalles (US) / us-west1
0.423 USD/h 0.7068 USD/h
Moncks Corner (US) / us-east1
0.423 USD/h 0.7068 USD/h
Mumbai (IN) / asia-south1
0.3747 USD/h 0.7358 USD/h
Eemshaven (NL) / europe-west4
0.4457 USD/h 0.7429 USD/h
Montréal (CA) / northamerica-northeast1
0.4414 USD/h 0.7783 USD/h
Toronto (CA) / northamerica-northeast2
0.4424 USD/h 0.7783 USD/h
St. Ghislain (BE) / europe-west1
0.467 USD/h 0.7783 USD/h
Las Vegas (US) / us-west4
0.4765 USD/h 0.7961 USD/h
London (GB) / europe-west2
0.4829 USD/h 0.8048 USD/h
Changhua County (TW) / asia-east1
0.4903 USD/h 0.8184 USD/h
Frankfurt (DE) / europe-west3
0.4973 USD/h 0.833 USD/h
Jurong West (SG) / asia-southeast1
0.5233 USD/h 0.872 USD/h
Seoul (KR) / asia-northeast3
0.3068 USD/h 0.9077 USD/h
Tokyo (JP) / asia-northeast1
0.4658 USD/h 0.9077 USD/h
Zurich (CH) / europe-west6
0.584 USD/h 0.9757 USD/h
Dammam (SA) / me-central2
0.638 USD/h 1.1309 USD/h

Processor

vCPUs
4
CPU Allocation
Dedicated
CPU Cores
2
CPU Speed
2.2 GHz
CPU Architecture
x86_64
CPU Manufacturer
Intel
CPU Family
Xeon
CPU L1D Cache
32 KiB
CPU L1D Cache Total
64 KiB
CPU L1I Cache
32 KiB
CPU L1I Cache Total
64 KiB
CPU L2 Cache
1 MiB
CPU L2 Cache Total
2 MiB
CPU L3 Cache
39 MiB
CPU L3 Cache Total
39 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, ss, ht, syscall, nx, pdpe1gb, rdtscp, lm, constant_tsc, rep_good, nopl, xtopology, nonstop_tsc, cpuid, tsc_known_freq, pni, pclmulqdq, ssse3, fma, cx16, pcid, sse4_1, sse4_2, x2apic, movbe, popcnt, aes, xsave, avx, f16c, rdrand, hypervisor, lahf_lm, abm, 3dnowprefetch, ssbd, ibrs, ibpb, stibp, ibrs_enhanced, fsgsbase, tsc_adjust, bmi1, hle, avx2, smep, bmi2, erms, invpcid, rtm, mpx, avx512f, avx512dq, rdseed, adx, smap, clflushopt, clwb, avx512cd, avx512bw, avx512vl, xsaveopt, xsavec, xgetbv1, xsaves, arat, avx512_vnni, md_clear, arch_capabilities
Ecpus
2.6
Scalability
65

System Resources and Accelerators

MEMORY
Memory Amount
16 GiB
Memory Amount Actual
16 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: 95.04.29.00.07, Clock rate: 2040)
STORAGE
Storage Size
0 GB
Storages
    NETWORK
    Inbound Traffic
    0 GB/month
    Outbound Traffic
    0 GB/month
    IPv4
    0

    CPU and System Topology

    Server Description

    A graphics-optimized virtual machine pairing a single NVIDIA L4 GPU with balanced host compute for machine learning inference and video processing tasks.

    GPU AcceleratedGeneral Purpose

    Google Cloud Platform g2-standard-4 is a graphics-optimized instance equipped with 1 NVIDIA L4 GPU containing 24 GB of VRAM, paired with 4 vCPUs and 16.0 GB of RAM. Powered by an Intel Xeon x86_64 processor running at 2.2 GHz across 2 physical cores, the host system provides basic compute support, while local storage is omitted. Performance data indicates modest host CPU capabilities, reflected in a PassMark CPU score of 3754.01 and a multi-core Geekbench score of 2294.0. However, GPU performance is strong for accelerated AI tasks, delivering 2296.42 tokens per second for 7B model prompt processing and 52.18 tokens per second for text generation. This balance minimizes host resource overhead, providing an efficient architecture for GPU-accelerated workloads such as medium-scale LLM inference, AI video encoding, and computer vision tasks.

    Economics

    Average Price per Region

    Prices per Zone

    Lowest Prices

    Performance

    Workload Profiles

    0.24Score
    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) 309,024 ops/sec 1,919,289 ops/sec 0.16 50.00% 50.00% -60.00%
    Redis RPS (pipeline=16, SET) 2,003,831 ops/sec 11,728,058 ops/sec 0.17 20.00% 20.00% -30.00%
    PassMark Memory Mark (composite) 1,836 2,416 0.75 10.00% 10.00% -2.80%
    Memory bandwidth (read, 16 MB ~ L3) 36,164 MB/sec 108,466 MB/sec 0.33 10.00% 10.00% -10.00%
    PassMark single-thread CPU 1,691 Mops/s 2,365 Mops/s 0.71 10.00% 10.00% -3.40%

    GPU Bandwidth and Latency

    Node (all GPUs)
    Node baseline directional bandwidth
    		6.1 GB/s
    Node GPU-to-host bandwidth
    		13.2 GB/s
    Node host-to-GPU bandwidth
    		12.4 GB/s
    Single GPU slot
    Per-GPU GPU-to-host bandwidth
    		13.2 GB/s
    Per-GPU host-to-GPU bandwidth
    		12.4 GB/s
    Per-GPU host-device latency
    		913 ns
    Benchmark Environment
    Driver Version
    		 535.183.01
    CUDA Runtime Version
    		 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: 1,051

    Geekbench Multi-Core

    Score: 2,294

    Passmark CPU Scores

    BENCHMARKSCORE
    Mark
    		 3,754
    Compression
    		 47,481
    Encryption
    		 1,449
    Extended Instructions
    		 2,901
    Floating Point Maths
    		 5,668
    Integer Maths
    		 10,529
    Physics
    		 508
    Prime Numbers
    		 23
    Single Threaded
    		 1,691
    String Sorting
    		 5,675

    Passmark Memory Scores

    BENCHMARKSCORE
    Memory Mark
    		 1,836
    Database Operations
    		 1,327
    Memory Latency
    		 52
    Memory Read Cached
    		 20,540
    Memory Read Uncached
    		 11,099
    Memory Write
    		 9,910

    Stress-ng Stressors

    CPU integerSingleMultiScaling
    atomic
    		 201.62 196.56 0.97×
    bitops
    		 91,792 210,848 2.30×
    branch
    		 130,739 395,906 3.03×
    funccall
    		 460.33 954.91 2.07×
    funcret
    		 9,646 27,349 2.84×
    hash
    		 107,104 310,978 2.90×
    intmath
    		 18,965 56,465 2.98×
    nop
    		 162,359 280,574 1.73×
    prime
    		 270,577 604,710 2.23×
    rotate
    		 82,273 225,989 2.75×
    str
    		 623,853 1,352,882 2.17×
    wcs
    		 398,224 869,150 2.18×
    CryptographySingleMultiScaling
    crypt
    		 2,072 4,036 1.95×
    CompressionSingleMultiScaling
    jpeg
    		 513.89 1,239 2.41×
    zlib
    		 48.38 131.92 2.73×
    HypervisorSingleMultiScaling
    clock
    		 26,583 60,252 2.27×
    clone
    		 286 514.75 1.80×
    context
    		 2,219 5,263 2.37×
    cpu-sched
    		 63.33 231.17 3.65×
    eventfd
    		 40,834 192,998 4.73×
    fault
    		 9,460 24,762 2.62×
    fork
    		 3,802 10,377 2.73×
    getrandom
    		 37,977 92,160 2.43×
    hrtimers
    		 208,578 318,190 1.53×
    mprotect
    		 517,316 558,603 1.08×
    mtx
    		 9,396,253 100,406,347 10.69×
    munmap
    		 2,507 6,458 2.58×
    null
    		 254,520 520,800 2.05×
    pipe
    		 808,743 1,278,626 1.58×
    priv-instr
    		 476,258 1,090,219 2.29×
    resched
    		 540,082 306,038 0.57×
    switch
    		 70,545 579,316 8.21×
    timermix
    		 164,612 473,547 2.88×
    tlb-shootdown
    		 6,647 8,823 1.33×
    tsc
    		 848,746 2,355,746 2.78×
    usersyscall
    		 332,681 744,932 2.24×
    vdso
    		 70,376,534 196,837,582 2.80×
    vfork
    		 26,447 63,337 2.39×
    x86syscall
    		 2,912,300 6,030,271 2.07×
    yield
    		 1,810,131 1,775,408 0.98×
    zero
    		 1,730,224 3,683,989 2.13×
    CPU float/SIMDSingleMultiScaling
    eigen
    		 13,702 33,638 2.45×
    expmath
    		 2,281 6,924 3.04×
    fma
    		 509,247 1,649,200 3.24×
    fp
    		 100.8 265.99 2.64×
    fractal
    		 4.2 12.4 2.95×
    hyperbolic
    		 2,351 7,398 3.15×
    logmath
    		 3,839 11,326 2.95×
    matrix
    		 2,586 8,322 3.22×
    matrix-3d
    		 600.86 1,044 1.74×
    monte-carlo
    		 118.5 287.11 2.42×
    mpfr
    		 1,027 2,199 2.14×
    powmath
    		 1,349 4,101 3.04×
    trig
    		 1,454 3,841 2.64×
    veccmp
    		 13,181 27,498 2.09×
    vecfp
    		 1,149 2,907 2.53×
    vecmath
    		 1,859 5,846 3.14×
    vecshuf
    		 480.32 1,027 2.14×
    vecwide
    		 22,021 47,548 2.16×
    AlgorithmsSingleMultiScaling
    bitonicsort
    		 6 14.4 2.40×
    bsearch
    		 203.58 465.22 2.29×
    bubblesort
    		 0.6 0.8 1.33×
    fibsearch
    		 207.26 472.3 2.28×
    heapsort
    		 7.4 17.6 2.38×
    hsearch
    		 4,262 10,733 2.52×
    insertionsort
    		 7 13.6 1.94×
    malloc
    		 1,952,287 7,741,630 3.97×
    mcontend
    		 189.58 803.96 4.24×
    mergesort
    		 19.8 53.2 2.69×
    qsort
    		 12.6 29 2.30×
    skiplist
    		 4,149 11,758 2.83×
    tree
    		 1.2 3.2 2.67×
    MemorySingleMultiScaling
    cache
    		 343,686 432,057 1.26×
    cachehammer
    		 365,186 1,319,282 3.61×
    cacheline
    		 24.54 99.01 4.03×
    l1cache
    		 193.35 415.71 2.15×
    memcpy
    		 295.08 632.27 2.14×
    memrate
    		 1.17 3.06 2.62×
    memthrash
    		 66.53 145.39 2.19×
    ptr-chase
    		 25,750,681 49,429,200 1.92×
    spinmem
    		 189.39 159.16 0.84×
    stream
    		 27.64 271.3 9.82×
    vm
    		 53,782 215,423 4.01×

    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
    g2-standard-8 832 GiB1
    g2-standard-12 1248 GiB1
    g2-standard-16 1664 GiB1
    g2-standard-24 2496 GiB2
    g2-standard-32 32128 GiB1
    g2-standard-48 48192 GiB4
    g2-standard-96 96384 GiB8

    Similar Servers

    INSTANCEVENDORvCPUsMEMORYAccelerators

    g2-standard-4 FAQs