logo

vcg-l40s-32c-375g-96vram by Vultr

vcg-l40s-32c-375g-96vram is a Cloud GPU (32 vCPUs, 375.0 GiB RAM, 2200 GB NVMe, 2xL40S 48 GiB VRAM) server offered by Vultr with 32 vCPUs, 375 GiB of memory and 2.2 TB of storage.
32 vCPU
375 GiB Memory
2.2 TB Storage
2 GPU
Spare SCore
62893
(All-cores)
2209
(Single-core)

Specifications

Server Metadata

Vendor ID
vultr
Name
vcg-l40s-32c-375g-96vram
Description
Cloud GPU (32 vCPUs, 375.0 GiB RAM, 2200 GB NVMe, 2xL40S 48 GiB VRAM)
Family
Cloud GPU
Hw Virt
Average Time To Start
135
Status
active
Observed At
2026-07-20T13:40:44.474754

Availability

REGION / IDSPOTONDEMAND

Processor

vCPUs
32
CPU Allocation
Shared
CPU Cores
32
CPU Speed
2 GHz
CPU Architecture
x86_64
CPU Manufacturer
Intel
CPU Family
Xeon
CPU Model
6448H
CPU L1D Cache
32 KiB
CPU L1D Cache Total
1 MiB
CPU L1I Cache
32 KiB
CPU L1I Cache Total
1 MiB
CPU L2 Cache
4 MiB
CPU L2 Cache Total
128 MiB
CPU L3 Cache
16 MiB
CPU L3 Cache Total
512 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, syscall, nx, pdpe1gb, rdtscp, lm, constant_tsc, arch_perfmon, rep_good, nopl, xtopology, cpuid, tsc_known_freq, pni, pclmulqdq, ssse3, fma, cx16, pdcm, pcid, sse4_1, sse4_2, x2apic, movbe, popcnt, tsc_deadline_timer, aes, xsave, avx, f16c, rdrand, hypervisor, lahf_lm, abm, cpuid_fault, ssbd, ibrs, ibpb, stibp, ibrs_enhanced, fsgsbase, tsc_adjust, bmi1, avx2, smep, bmi2, erms, invpcid, avx512f, avx512dq, avx512ifma, clflushopt, clwb, avx512cd, sha_ni, avx512bw, avx512vl, xsaveopt, xsaves, avx512_bf16, wbnoinvd, arat, avx512vbmi, umip, pku, ospke, avx512_vbmi2, gfni, vaes, vpclmulqdq, avx512_vnni, avx512_bitalg, avx512_vpopcntdq, la57, rdpid, cldemote, movdiri, movdir64b, fsrm, md_clear, serialize, tsxldtrk, arch_capabilities
Ecpus
28.5
Scalability
89.06

System Resources and Accelerators

MEMORY
Memory Amount
375 GiB
Memory Amount Actual
375 GiB
GPU
GPU Count
2
GPU Memory Min
48 GiB
GPU Memory Total
96 GiB
GPU Manufacturer
NVIDIA
GPU Family
Ada Lovelace
GPU Model
L40S
GPUs
  • NVIDIA Ada Lovelace NVIDIA L40S (Memory amount: 49140, Firmware version: 580.95.05, BIOS version: 95.02.66.00.02, Clock rate: 2520)
  • NVIDIA Ada Lovelace NVIDIA L40S (Memory amount: 49140, Firmware version: 580.95.05, BIOS version: 95.02.66.00.02, Clock rate: 2520)
STORAGE
Storage Size
2200 GB
Storage Type
nvme ssd
Storages
    NETWORK
    Inbound Traffic
    0 GB/month
    Outbound Traffic
    10240 GB/month
    IPv4
    1

    CPU and System Topology

    Server Description

    A high-density GPU server combining dual NVIDIA L40S accelerators, 375 GB RAM, and NVMe storage for demanding AI inference and database workloads.

    GPU AcceleratedMemory OptimizedStorage & Database

    Vultr vcg-l40s-32c-375g-96vram is a high-density Cloud GPU server powered by an Intel Xeon 6448H processor with 32 physical cores and 32 vCPUs. It features 375.0 GB of system memory and a 2200 GB NVMe SSD. The server's primary compute acceleration comes from two NVIDIA L40S Ada Lovelace GPUs, providing a total of 96 GB of VRAM. Benchmark results show top-tier performance in LLM inference, with prompt processing speeds of 6,275.36 tokens per second on a 7B model, and strong multi-core CPU capabilities. While large-block memory write bandwidth is average, the system excels in memory read bandwidth and database operations, registering 70.17 million operations per second in Redis SET tests. This hardware configuration is highly suited for machine learning inference, data science, and memory-intensive database workloads.

    Economics

    Average Price per Region

    Prices per Zone

    Lowest Prices

    Performance

    Workload Profiles

    2.83Score
    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) 6,767,958 ops/sec 1,895,870 ops/sec 3.57 50.00% 50.00% +88.90%
    Redis RPS (pipeline=16, SET) 44,545,708 ops/sec 11,503,938 ops/sec 3.87 20.00% 20.00% +31.10%
    PassMark Memory Mark (composite) 2,458 2,410 1.02 10.00% 10.00% +0.20%
    Memory bandwidth (read, 16 MB ~ L3) 285,063 MB/sec 107,331 MB/sec 2.66 10.00% 10.00% +10.30%
    PassMark single-thread CPU 3,319 Mops/s 2,363 Mops/s 1.4 10.00% 10.00% +3.42%

    Memory Bandwidth

    Compression

    OpenSSL

    Passmark CPU Scores

    BENCHMARKSCORE
    Mark
    		64769
    Compression
    		790586
    Encryption
    		44502
    Extended Instructions
    		83840
    Floating Point Maths
    		209427
    Integer Maths
    		179238
    Physics
    		9625
    Prime Numbers
    		611
    Single Threaded
    		3319
    String Sorting
    		106349

    Passmark Memory Scores

    BENCHMARKSCORE
    Memory Mark
    		2458
    Database Operations
    		28858
    Memory Latency
    		67
    Memory Read Cached
    		30089
    Memory Read Uncached
    		10218
    Memory Write
    		8797

    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

    INSTANCEvCPUsMEMORYGPUs
    vcg-a40-1c-5g-2vram 15 GiB1⁄24
    vcg-a40-2c-10g-4vram 210 GiB1⁄12
    vcg-a16-2c-8g-2vram 28 GiB⅛
    vcg-a16-2c-16g-4vram 216 GiB¼
    vcg-a16-3c-32g-8vram 332 GiB½
    vcg-a40-4c-20g-8vram 420 GiB⅙
    vcg-a40-6c-30g-12vram 630 GiB¼

    Similar Servers

    INSTANCEVENDORvCPUsMEMORYGPUs

    vcg-l40s-32c-375g-96vram FAQs