AWS
AWS offers G4 instances optimized for machine learning inference and graphics-intensive applications. These instances provide cost-effective options using NVIDIA or AMD GPUs, specifically tailored for tasks like AI model deployment, gaming, and high-end graphics rendering.
AWS
Provider Profile
Founded
2006
Headquarters
Seattle, WA, USA
Pricing Model
Pay-as-you-go with various pricing tiers depending on instance types
Technical Specification
Target Audience
- Enterprise
- academics
- researchers in machine learning and graphics-intensive computing
GPU Clusters & Offerings
- NVIDIA T4 GPUs
- AMD Radeon Pro V520 GPUs
- Machine learning inference and training
- Graphics-intensive applications like game streaming and 3D rendering
Network Fabric
- 100 Gbps Ethernet
- Elastic Fabric Adapter
Connectivity Bandwidth
Up to 100 Gbps
Storage Architecture
- Local NVMe-based SSD
- Amazon EBS (Elastic Block Store)
- Amazon S3 (Simple Storage Service)
Compute Framework Compatibility
- TensorFlow
- PyTorch
- Apache MXNet
- CUDA
- CuDNN
Resource Orchestration
- Amazon ECS
- Amazon EKS
- Kubernetes
Security Infrastructure
Comprehensive vulnerability assessments and regular security audits
Developer Interface & APIs
- AWS CLI
- SDKs for popular languages
- RESTful API
- AWS CloudFormation
Support Operations
- 24/7 technical support
- Documentation
- Community forums
- Premium Support tiers (additional cost)
Resource Availability
GA (Generally Available)
Datacenter Locations
Regulatory Compliance
- ISO 27018
- ISO 27001
- ISO 22301
- PCI DSS
- SOC 1
- SOC 2
- SOC 3
Key Platform Features
- Local NVMe-based SSD storage
- Up to 100 Gbps networking
- Elastic Fabric Adapter (EFA)
- CUDA, CuDNN, and NVENC for NVIDIA
- DirectX, OpenGL, and Vulkan for AMD
- Bare metal capabilities
Last Audit: February 2026