g6.48xlarge by Amazon Web Services
(All-cores)
(Single-core)
Specifications
Server Metadata
Vendor ID | aws |
Name | g6.48xlarge |
Description | Graphics intensive Gen6 48xlarge |
Family | g6 |
Hw Virt | |
Status | active |
Observed At | 2026-10-29T17:19:31.530431 |
Availability
| REGION / ID | SPOT | ONDEMAND |
|---|---|---|
Northern Virgina (US) / us-east-1 | 7.19106 USD/h | 13.3504 USD/h |
Oregon (US) / us-west-2 | 6.476525 USD/h | 13.3504 USD/h |
Ohio (US) / us-east-2 | 6.9912 USD/h | 13.3504 USD/h |
Aragón (ES) / eu-south-2 | 5.144733 USD/h | 14.067 USD/h |
Stockholm (SE) / eu-north-1 | 2.74645 USD/h | 14.1599 USD/h |
Quebec (CA) / ca-central-1 | 14.8235 USD/h | 14.8235 USD/h |
Hyderabad (IN) / ap-south-2 | 7.2036 USD/h | 16.0311 USD/h |
Mumbai (IN) / ap-south-1 | 16.0311 USD/h | 16.0311 USD/h |
United Arab Emirates / me-central-1 | 6.7111 USD/h | 16.3934 USD/h |
Seoul (KR) / ap-northeast-2 | 7.7253 USD/h | 16.4159 USD/h |
Frankfurt (DE) / eu-central-1 | 5.533067 USD/h | 16.6946 USD/h |
(MY) / ap-southeast-5 | 2.2725 USD/h | 16.8174 USD/h |
London (GB) / eu-west-2 | 8.737667 USD/h | 16.9468 USD/h |
Paris (FR) / eu-west-3 | 5.20495 USD/h | 16.9468 USD/h |
Sydney (AU) / ap-southeast-2 | 12.2294 USD/h | 17.3582 USD/h |
Zurich (CH) / eu-central-2 | 5.79195 USD/h | 17.9985 USD/h |
Tokyo (JP) / ap-northeast-1 | 8.81495 USD/h | 19.3621 USD/h |
Sao Paulo (BR) / sa-east-1 | 4.06745 USD/h | 22.693 USD/h |
Processor
vCPUs | 192 |
Hypervisor | nitro |
CPU Allocation | Dedicated |
CPU Cores | 96 |
CPU Speed | 3.4 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 | 96 |
Scalability | 50 |
System Resources and Accelerators
| MEMORY | |
|---|---|
Memory Amount | 768 GiB |
Memory Amount Actual | 768 GiB |
Memory Generation | DDR4 |
Memory Speed | 3200 Mhz |
| ACCELERATORS | |
|---|---|
Accelerator Count | 8 |
Accelerator Memory Min | 22 GiB |
Accelerator Memory Total | 179 GiB |
Accelerator Type | GPU |
Accelerator Manufacturer | NVIDIA |
Accelerator Family | Ada Lovelace |
Accelerator Model | L4 |
Accelerators |
|
| STORAGE | |
|---|---|
Storage Size | 7520 GB |
Storage Type | nvme ssd |
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 |
CPU and System Topology
Server Description
A highly parallelized x86 server featuring eight Ada Lovelace GPUs, massive NVMe storage, and top-tier multi-core and LLM inference performance.
Amazon Web Services g6.48xlarge is a GPU-accelerated server utilizing the AWS Nitro hypervisor. It features 192 vCPUs from AMD EPYC 7R13 processors running at 3.4 GHz, paired with 768.0 GB of DDR4 memory. The instance is equipped with eight NVIDIA L4 Ada Lovelace GPUs providing 179 GB of total VRAM, 7520 GB of local NVMe SSD storage, and a 100 Gbps baseline network bandwidth. In performance testing, the server achieves top-tier results in multi-core CPU benchmarks, ray tracing, and LLM inference for small and medium models, though single-core performance remains average. This hardware configuration is designed for parallelized workloads, including large-scale LLM inference, graphics rendering, and high-throughput database operations.
Economics
Performance
GPU Bandwidth and Latency
| Node (all GPUs) | |
|---|---|
Node baseline directional bandwidth | 88.6 GB/s |
Node GPU-to-host bandwidth | 105.7 GB/s |
Node host-to-GPU bandwidth | 107.7 GB/s |
| Single GPU slot | |
|---|---|
Per-GPU GPU-to-host bandwidth | 13.2 GB/s |
Per-GPU host-to-GPU bandwidth | 13.5 GB/s |
Per-GPU host-device latency | 806 ns |
| Benchmark Environment | |
|---|---|
Driver Version | 535.183.01 |
CUDA Version (driver) | 12.6 |
nvbandwidth Version | v0.5 |
GPU Count | 8 |
P2P Supported | No |
| Efficiency | |
|---|---|
SM/CE copy efficiency | 1.00 |
Memory Bandwidth
Compression
OpenSSL
Geekbench Single-Core
Geekbench Multi-Core
Passmark CPU Scores
| BENCHMARK | SCORE |
|---|---|
Mark | 109,870 |
Compression | 2,608,148 |
Encryption | 193,116 |
Extended Instructions | 133,785 |
Floating Point Maths | 434,975 |
Integer Maths | 818,923 |
Physics | 14,964 |
Prime Numbers | 990 |
Single Threaded | 2,644 |
String Sorting | 337,479 |
Passmark Memory Scores
| BENCHMARK | SCORE |
|---|---|
Memory Mark | 2,806 |
Database Operations | 29,517 |
Memory Latency | 68 |
Memory Read Cached | 26,363 |
Memory Read Uncached | 16,755 |
Memory Write | 17,249 |
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
| INSTANCE | vCPUs | MEMORY | Accelerators |
|---|---|---|---|
| g6.xlarge | 4 | 16 GiB | 1 |
| g6.2xlarge | 8 | 32 GiB | 1 |
| g6.4xlarge | 16 | 64 GiB | 1 |
| g6.8xlarge | 32 | 128 GiB | 1 |
| g6.12xlarge | 48 | 192 GiB | 4 |
| g6.16xlarge | 64 | 256 GiB | 1 |
| g6.24xlarge | 96 | 384 GiB | 4 |
Similar Servers
| INSTANCE | VENDOR | vCPUs | MEMORY | Accelerators |
|---|