g2-standard-12 by Google Cloud Platform

g2-standard-12 is a Graphics Optimized: 1 NVIDIA L4 GPU, 12 vCPUs, 48GB RAM server offered by Google Cloud Platform with 12 vCPUs, 48 GiB of memory and 0 GB of storage. The pricing starts at 0.4331 USD per hour.
12 vCPU
48 GiB Memory
1 GPU
Spare SCore
9,484
(All-cores)
1,198
(Single-core)

Specifications

Server Metadata

Vendor ID
gcp
Server ID
1600012
Name
g2-standard-12
Description
Graphics Optimized: 1 NVIDIA L4 GPU, 12 vCPUs, 48GB RAM
Family
g2
Hw Virt
Average Time To Start
170
Status
active
Observed At
2026-11-09T02:35:39.032739

Availability

REGION / IDSPOTONDEMAND
Ashburn (US) / us-east4
0.5307 USD/h 0.9971 USD/h
Council Bluffs (US) / us-central1
0.5986 USD/h 1.0004 USD/h
The Dalles (US) / us-west1
0.5986 USD/h 1.0004 USD/h
Moncks Corner (US) / us-east1
0.5986 USD/h 1.0004 USD/h
Mumbai (IN) / asia-south1
0.5289 USD/h 1.0414 USD/h
Eemshaven (NL) / europe-west4
0.6308 USD/h 1.0515 USD/h
Montréal (CA) / northamerica-northeast1
0.6559 USD/h 1.1016 USD/h
Toronto (CA) / northamerica-northeast2
0.6574 USD/h 1.1016 USD/h
St. Ghislain (BE) / europe-west1
0.661 USD/h 1.1016 USD/h
Las Vegas (US) / us-west4
0.6744 USD/h 1.1268 USD/h
London (GB) / europe-west2
0.6834 USD/h 1.1391 USD/h
Changhua County (TW) / asia-east1
0.694 USD/h 1.1584 USD/h
Frankfurt (DE) / europe-west3
0.7039 USD/h 1.179 USD/h
Jurong West (SG) / asia-southeast1
0.7406 USD/h 1.2342 USD/h
Seoul (KR) / asia-northeast3
0.4331 USD/h 1.2847 USD/h
Tokyo (JP) / asia-northeast1
0.6889 USD/h 1.2847 USD/h
Zurich (CH) / europe-west6
0.8141 USD/h 1.3604 USD/h
Dammam (SA) / me-central2
0.9009 USD/h 1.6007 USD/h

Processor

vCPUs
12
CPU Allocation
Dedicated
CPU Cores
6
CPU Speed
2.2 GHz
CPU Architecture
x86_64
CPU Manufacturer
Intel
CPU Family
Xeon
CPU L1D Cache
32 KiB
CPU L1D Cache Total
192 KiB
CPU L1I Cache
32 KiB
CPU L1I Cache Total
192 KiB
CPU L2 Cache
1 MiB
CPU L2 Cache Total
6 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
7.9
Scalability
65.83

System Resources and Accelerators

MEMORY
Memory Amount
48 GiB
Memory Amount Actual
48 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

    An accelerator-focused configuration pairing a dedicated graphics processor with balanced system memory for efficient machine learning inference.

    GPU AcceleratedGeneral Purpose

    Google Cloud Platform g2-standard-12 is a graphics-optimized server featuring a dedicated NVIDIA L4 GPU with 24 GB of VRAM alongside an Intel Xeon x86_64 processor with 12 vCPUs and 48.0 GB of system memory. It does not include local storage. Performance benchmarks reveal weak single-core CPU execution, while multi-core operations and memory bandwidth perform in the average tier. However, the server excels in GPU-driven tasks, achieving top-tier results in LLM inference for small and medium models. This configuration offers a qualitatively cost-efficient entry point for GPU acceleration, though users must accept the tradeoff of lower single-threaded CPU speeds and the necessity of external storage. It is best suited for small-to-medium LLM inference, graphics rendering, and machine learning workloads.

    Economics

    Average Price per Region

    Prices per Zone

    Lowest Prices

    Performance

    Workload Profiles

    0.56Score
    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) 924,163 ops/sec 1,922,940 ops/sec 0.48 50.00% 50.00% -31.00%
    Redis RPS (pipeline=16, SET) 5,796,044 ops/sec 11,797,257 ops/sec 0.48 20.00% 20.00% -14.00%
    PassMark Memory Mark (composite) 2,409 2,406 1 10.00% 10.00% 0.00%
    Memory bandwidth (read, 16 MB ~ L3) 75,077 MB/sec 108,657 MB/sec 0.68 10.00% 10.00% -3.80%
    PassMark single-thread CPU 1,695 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
    		12.3 GB/s
    Single GPU slot
    Per-GPU GPU-to-host bandwidth
    		13.2 GB/s
    Per-GPU host-to-GPU bandwidth
    		12.3 GB/s
    Per-GPU host-device latency
    		910 ns
    Benchmark Environment
    Driver Version
    		 535.183.01
    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: 1,063

    Geekbench Multi-Core

    Score: 5,383

    Passmark CPU Scores

    BENCHMARKSCORE
    Mark
    		 10,478
    Compression
    		 139,689
    Encryption
    		 4,346
    Extended Instructions
    		 8,971
    Floating Point Maths
    		 17,053
    Integer Maths
    		 31,706
    Physics
    		 1,179
    Prime Numbers
    		 61
    Single Threaded
    		 1,695
    String Sorting
    		 16,728

    Passmark Memory Scores

    BENCHMARKSCORE
    Memory Mark
    		 2,409
    Database Operations
    		 3,900
    Memory Latency
    		 52
    Memory Read Cached
    		 20,776
    Memory Read Uncached
    		 10,477
    Memory Write
    		 9,783

    Stress-ng Stressors

    CPU integerSingleMultiScaling
    atomic
    		 197.08 186.22 0.94×
    bitops
    		 92,340 634,841 6.88×
    branch
    		 131,285 1,193,234 9.09×
    funccall
    		 462.54 2,860 6.18×
    funcret
    		 9,590 82,470 8.60×
    hash
    		 106,943 929,130 8.69×
    intmath
    		 18,953 170,001 8.97×
    nop
    		 163,120 844,204 5.18×
    prime
    		 271,728 1,819,719 6.70×
    rotate
    		 82,321 677,713 8.23×
    str
    		 642,859 4,182,300 6.51×
    wcs
    		 401,819 2,574,355 6.41×
    CryptographySingleMultiScaling
    crypt
    		 2,082 12,152 5.84×
    CompressionSingleMultiScaling
    jpeg
    		 512.37 3,652 7.13×
    zlib
    		 47.03 384.69 8.18×
    HypervisorSingleMultiScaling
    clock
    		 26,648 157,944 5.93×
    clone
    		 304.56 1,502 4.93×
    context
    		 2,242 15,541 6.93×
    cpu-sched
    		 63.27 792.82 12.53×
    eventfd
    		 38,488 498,575 12.95×
    fault
    		 9,562 72,641 7.60×
    fork
    		 3,648 24,826 6.81×
    getrandom
    		 38,156 277,970 7.29×
    hrtimers
    		 250,289 1,169,271 4.67×
    mprotect
    		 908,668 1,671,666 1.84×
    mtx
    		 8,970,101 262,348,172 29.25×
    munmap
    		 2,416 18,021 7.46×
    null
    		 257,006 1,567,737 6.10×
    pipe
    		 840,652 3,652,766 4.35×
    priv-instr
    		 477,117 3,240,789 6.79×
    resched
    		 1,975,602 758,237 0.38×
    switch
    		 75,831 1,762,831 23.25×
    timermix
    		 167,175 1,420,418 8.50×
    tlb-shootdown
    		 10,273 22,305 2.17×
    tsc
    		 849,581 7,097,826 8.35×
    usersyscall
    		 331,219 2,226,962 6.72×
    vdso
    		 70,540,806 592,782,576 8.40×
    vfork
    		 26,708 136,064 5.09×
    x86syscall
    		 2,894,832 17,946,817 6.20×
    yield
    		 5,977,509 6,204,107 1.04×
    zero
    		 1,723,346 11,035,609 6.40×
    CPU float/SIMDSingleMultiScaling
    eigen
    		 13,744 99,723 7.26×
    expmath
    		 2,286 20,912 9.15×
    fma
    		 510,833 4,979,189 9.75×
    fp
    		 110.4 788.17 7.14×
    fractal
    		 4.2 37 8.81×
    hyperbolic
    		 2,377 22,299 9.38×
    logmath
    		 3,864 34,106 8.83×
    matrix
    		 2,593 24,974 9.63×
    matrix-3d
    		 610.59 2,518 4.12×
    monte-carlo
    		 118.56 796.35 6.72×
    mpfr
    		 1,032 6,640 6.44×
    powmath
    		 1,350 12,353 9.15×
    trig
    		 1,458 11,577 7.94×
    veccmp
    		 13,284 82,664 6.22×
    vecfp
    		 1,150 8,614 7.49×
    vecmath
    		 1,862 17,613 9.46×
    vecshuf
    		 481.23 3,095 6.43×
    vecwide
    		 22,136 143,179 6.47×
    AlgorithmsSingleMultiScaling
    bitonicsort
    		 6.2 43 6.94×
    bsearch
    		 203.86 1,400 6.87×
    bubblesort
    		 0.6 2.4 4.00×
    fibsearch
    		 208.41 1,430 6.86×
    heapsort
    		 7.4 52.8 7.14×
    hsearch
    		 4,266 32,161 7.54×
    insertionsort
    		 7 40.8 5.83×
    malloc
    		 1,985,117 19,080,239 9.61×
    mcontend
    		 124.65 2,383 19.12×
    mergesort
    		 19.8 160.59 8.11×
    qsort
    		 12.6 86.8 6.89×
    skiplist
    		 4,159 35,245 8.47×
    tree
    		 1.2 7.2 6.00×
    MemorySingleMultiScaling
    cache
    		 345,144 629,890 1.83×
    cachehammer
    		 361,308 3,815,627 10.56×
    cacheline
    		 38.87 365.99 9.42×
    l1cache
    		 185.64 1,257 6.77×
    memcpy
    		 296.31 1,905 6.43×
    memrate
    		 1.17 6.86 5.86×
    memthrash
    		 105.06 417.41 3.97×
    ptr-chase
    		 25,749,712 125,017,996 4.86×
    spinmem
    		 190.59 442.14 2.32×
    stream
    		 27.84 2,372 85.19×
    vm
    		 52,707 292,432 5.55×

    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-4 416 GiB1
    g2-standard-8 832 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-12 FAQs