AWS EC2 Instance GPU Types

AWS EC2 GPU Instance Comparison

FeatureAmazon EC2 G5Amazon EC2 G6Amazon EC2 G7
Processor2nd-generation AMD EPYC processorsNewer-generation processors optimized for GPU workloadsCustom Intel Xeon 6 processors
GPUUp to 8 NVIDIA A10G Tensor Core GPUsNVIDIA GPUs with enhanced memory bandwidthUp to 8 NVIDIA RTX PRO 4500 Blackwell Server Edition GPUs
GPU Memory24 GB per GPU24 GB per GPU32 GB per GPU
AI PerformanceCost-effective inference and small-to-medium model trainingLower-latency inference and mid-scale AI deploymentUp to 4.6× higher AI inference performance than G6
Graphics PerformanceSuitable for 3D rendering and remote graphics workstationsOptimized for complex graphics and video workloadsUp to 2.1× higher graphics performance than G6
Network BandwidthUp to 100 GbpsUp to 100 GbpsUp to 700 Gbps with EFA—7× G6
Local NVMe StorageUp to 7.6 TBUp to 7.52 TBUp to 7.6 TB
Video CapabilitiesGeneral-purpose graphics and video processingComplex video editing and transcoding9th-generation NVENC and 6th-generation NVDEC with 4:2:2 support
Video Stream PerformanceStandard A10G video capabilitiesEnhanced video processingUp to 1.6× more concurrent video streams than G6
Primary Use CasesML inference, small-to-medium training, 3D rendering, and remote workstationsReal-time processing, video editing/transcoding, and mid-scale AI modelsHigh-performance AI inference, advanced graphics, large-scale video processing, and distributed GPU workloads
Best FitCost-sensitive and established GPU workloadsBalanced inference, graphics, and video workloadsHighest performance, network throughput, and GPU memory among the three generations

Summary: G5 is the cost-effective option for general GPU workloads, G6 improves latency and video processing, and G7 provides the strongest AI, graphics, networking, and video performance.

Read more