{ "features_resolved": { "snapshot_any": true, "snapshot_mem": false, "fork": false, "pause_resume": true, "persistent_disk": true, "whole_disk": true, "persist_scope": null, "volumes": true, "auto_stop_idle": false, "long_sessions": true, "custom_image": true, "image_snapshot": true, "image_oci": false, "vm_isolation": true, "docker_inside": true, "nested_virt": true, "root": true, "systemd": true, "browser": false, "desktop": false, "computer_use": false, "code_interpreter": false, "browser_control": null, "desktop_control": null, "browser_and_desktop": false, "anti_bot": false, "captcha_solving": false, "residential_ip": false, "agent_harnesses": false, "ssh": true, "public_ipv4": true, "inbound_https": false, "custom_domain": true, "raw_tcp_inbound": true, "egress_allowlist": true, "open_internet": true, "static_egress_ip": true, "private_network": true, "self_host": false, "byoc": false, "open_source": false, "soc2": true, "hipaa": true, "sso": true, "gpu": true, "arm64": true, "windows": true, "macos": false, "lang_python": true, "lang_node": false, "lang_go": null, "lang_rust": null, "lang_java": null, "gpu_desktop": false, "credential_injection": null, "egress_http_rules": true, "wake_on_request": false, "live_resize": false, "fork_running": false, "memory_snapshot_fork": false, "webhooks": null, "scheduled_wakeups": null, "audit_logs": null, "eu_data_residency": true, "zero_data_retention": null, "scoped_api_keys": null, "spend_limits": null, "mcp_server": false, "usage_api": true, "snapshot_auto": true, "snapshot_on_demand": true, "harness_api": false, "own_agent_api": false, "fast_boot": null, "strong_isolation": true, "ingress_rules": true, "guest_firewall": true, "secret_proxy": false, "secret_proxy_any": false, "volume_attach": true, "volume_shared": true, "sidecar": null, "fuse": true, "kernel_features": true, "burst": null }, "company": { "id": "aws-ec2", "checked": "2026-09-30", "company": "Amazon Web Services, Inc. (a subsidiary of Amazon.com, Inc.)", "ownership": { "value": "subsidiary", "parent": "Amazon.com, Inc.", "source": "https://en.wikipedia.org/wiki/Amazon_Web_Services", "quote": "Amazon Web Services, Inc. (AWS) is a subsidiary of Amazon that provides on-demand cloud computing platforms and APIs to individuals, companies, and governments, on a metered, pay-as-you-go basis." }, "stage": { "value": "subsidiary", "source": "https://en.wikipedia.org/wiki/Amazon_Web_Services", "quote": "Amazon Web Services, Inc. (AWS) is a subsidiary of Amazon", "confidence": "high", "note": "EC2 is a service within AWS, a wholly-owned division of publicly traded Amazon.com, Inc. (NASDAQ: AMZN). Not independently funded; no accelerator/VC history applies." }, "accelerators": [], "total_funding_usd": { "value": null, "status": "unknown" }, "last_round": { "type": null, "amount_usd": null, "date": null, "leads": [], "status": "unknown", "source": "", "quote": "" }, "investors": [], "portfolio_listings": [], "revenue": { "value_usd": 128700000000, "kind": "annual revenue", "as_of": "2025", "status": "confirmed", "basis": "AWS segment revenue as reported in Amazon.com's Q4/FY2025 earnings release, cited by Wikipedia's AWS infobox.", "source": "https://ir.aboutamazon.com/news-release/news-release-details/2026/Amazon-com-Announces-Fourth-Quarter-Results/", "quote": "Revenue US$128.7 billion (2025)" }, "employees": { "value": null, "source": "" }, "note": "AWS EC2 is a product line inside AWS, itself a subsidiary of Amazon.com, Inc. (public company, NASDAQ: AMZN). Figures above are for the whole AWS segment, not EC2 alone; EC2-specific revenue is not broken out by Amazon. Amazon.com as a whole reported ~$638B total net sales for FY2024 (not used here since AWS segment revenue is more relevant)." }, "id": "aws-ec2", "name": "AWS EC2 (reference VMs)", "url": "https://aws.amazon.com/ec2/pricing/on-demand/", "category": "hyperscaler", "isolation": "vm", "modes": [ { "key": "on-demand", "label": "x86 on-demand c7i/m7i (us-east-1)", "flags": [], "os": [ "linux", "windows" ], "regions": [ "us" ], "gpu_only": false, "features": { "isolation": "vm", "nested_virt": false }, "pricing": "sizes", "vcpu_h": null, "ram_gib_h": null, "cpu_basis": "alloc", "ram_basis": "alloc", "active_floor": 0, "ram_per_vcpu": [ null, null ], "ram_fixed_per_vcpu": null, "vcpu_options": null, "min_vcpu": null, "max_vcpu": null, "max_ram_gib": null, "sizes": [ { "name": "c7i.large", "vcpu": 2, "ram_gib": 4, "disk_gib": 0, "hour": 0.08925, "month_cap": null }, { "name": "m7i.large", "vcpu": 2, "ram_gib": 8, "disk_gib": 0, "hour": 0.1008, "month_cap": null }, { "name": "c7i.xlarge", "vcpu": 4, "ram_gib": 8, "disk_gib": 0, "hour": 0.1785, "month_cap": null }, { "name": "m7i.xlarge", "vcpu": 4, "ram_gib": 16, "disk_gib": 0, "hour": 0.2016, "month_cap": null }, { "name": "c7i.2xlarge", "vcpu": 8, "ram_gib": 16, "disk_gib": 0, "hour": 0.357, "month_cap": null }, { "name": "m7i.2xlarge", "vcpu": 8, "ram_gib": 32, "disk_gib": 0, "hour": 0.4032, "month_cap": null }, { "name": "c7i.4xlarge", "vcpu": 16, "ram_gib": 32, "disk_gib": 0, "hour": 0.714, "month_cap": null }, { "name": "m7i.4xlarge", "vcpu": 16, "ram_gib": 64, "disk_gib": 0, "hour": 0.8064, "month_cap": null }, { "name": "c7i.8xlarge", "vcpu": 32, "ram_gib": 64, "disk_gib": 0, "hour": 1.428, "month_cap": null }, { "name": "m7i.8xlarge", "vcpu": 32, "ram_gib": 128, "disk_gib": 0, "hour": 1.6128, "month_cap": null } ], "pool_tiers": null, "requires_always_on": false, "always_on_month_per_instance": null, "commit_note": null, "min_billed_seconds": 60, "granularity_s": 1, "start_fee": 0, "boot_overhead_s": null, "multiplier": 1, "windows_vcpu_h": 0.046, "gpu": null, "note": "Price List API (AmazonEC2 us-east-1, 2026-09-25), Linux, shared tenancy. Windows licence-included adds $0.184/h per 4 vCPU = $0.046/vCPU-h (c7i.xlarge $0.3625).", "product_class": "vm", "cpu_class": "dedicated" }, { "key": "on-demand-eu", "label": "x86 on-demand c7i/m7i (eu-north-1, cheapest EU checked)", "flags": [], "os": [ "linux", "windows" ], "regions": [ "eu" ], "gpu_only": false, "features": { "isolation": "vm", "nested_virt": false }, "pricing": "sizes", "vcpu_h": null, "ram_gib_h": null, "cpu_basis": "alloc", "ram_basis": "alloc", "active_floor": 0, "ram_per_vcpu": [ null, null ], "ram_fixed_per_vcpu": null, "vcpu_options": null, "min_vcpu": null, "max_vcpu": null, "max_ram_gib": null, "sizes": [ { "name": "c7i.large", "vcpu": 2, "ram_gib": 4, "disk_gib": 0, "hour": 0.09555, "month_cap": null }, { "name": "m7i.large", "vcpu": 2, "ram_gib": 8, "disk_gib": 0, "hour": 0.1071, "month_cap": null }, { "name": "c7i.xlarge", "vcpu": 4, "ram_gib": 8, "disk_gib": 0, "hour": 0.1911, "month_cap": null }, { "name": "m7i.xlarge", "vcpu": 4, "ram_gib": 16, "disk_gib": 0, "hour": 0.2142, "month_cap": null }, { "name": "c7i.2xlarge", "vcpu": 8, "ram_gib": 16, "disk_gib": 0, "hour": 0.3822, "month_cap": null }, { "name": "m7i.2xlarge", "vcpu": 8, "ram_gib": 32, "disk_gib": 0, "hour": 0.4284, "month_cap": null }, { "name": "c7i.4xlarge", "vcpu": 16, "ram_gib": 32, "disk_gib": 0, "hour": 0.7644, "month_cap": null }, { "name": "m7i.4xlarge", "vcpu": 16, "ram_gib": 64, "disk_gib": 0, "hour": 0.8568, "month_cap": null }, { "name": "c7i.8xlarge", "vcpu": 32, "ram_gib": 64, "disk_gib": 0, "hour": 1.5288, "month_cap": null }, { "name": "m7i.8xlarge", "vcpu": 32, "ram_gib": 128, "disk_gib": 0, "hour": 1.7136, "month_cap": null } ], "pool_tiers": null, "requires_always_on": false, "always_on_month_per_instance": null, "commit_note": null, "min_billed_seconds": 60, "granularity_s": 1, "start_fee": 0, "boot_overhead_s": null, "multiplier": 1, "windows_vcpu_h": 0.046, "gpu": null, "note": "eu-north-1 Stockholm; eu-west-1 c7i.xlarge $0.19152, eu-central-1 $0.2037.", "product_class": "vm", "cpu_class": "dedicated" }, { "key": "on-demand-asia", "label": "x86 on-demand c7i/m7i (ap-south-1, cheapest Asia checked)", "flags": [], "os": [ "linux", "windows" ], "regions": [ "asia" ], "gpu_only": false, "features": { "isolation": "vm", "nested_virt": false }, "pricing": "sizes", "vcpu_h": null, "ram_gib_h": null, "cpu_basis": "alloc", "ram_basis": "alloc", "active_floor": 0, "ram_per_vcpu": [ null, null ], "ram_fixed_per_vcpu": null, "vcpu_options": null, "min_vcpu": null, "max_vcpu": null, "max_ram_gib": null, "sizes": [ { "name": "c7i.large", "vcpu": 2, "ram_gib": 4, "disk_gib": 0, "hour": 0.08925, "month_cap": null }, { "name": "m7i.large", "vcpu": 2, "ram_gib": 8, "disk_gib": 0, "hour": 0.10605, "month_cap": null }, { "name": "c7i.xlarge", "vcpu": 4, "ram_gib": 8, "disk_gib": 0, "hour": 0.1785, "month_cap": null }, { "name": "m7i.xlarge", "vcpu": 4, "ram_gib": 16, "disk_gib": 0, "hour": 0.2121, "month_cap": null }, { "name": "c7i.2xlarge", "vcpu": 8, "ram_gib": 16, "disk_gib": 0, "hour": 0.357, "month_cap": null }, { "name": "m7i.2xlarge", "vcpu": 8, "ram_gib": 32, "disk_gib": 0, "hour": 0.4242, "month_cap": null }, { "name": "c7i.4xlarge", "vcpu": 16, "ram_gib": 32, "disk_gib": 0, "hour": 0.714, "month_cap": null }, { "name": "m7i.4xlarge", "vcpu": 16, "ram_gib": 64, "disk_gib": 0, "hour": 0.8484, "month_cap": null }, { "name": "c7i.8xlarge", "vcpu": 32, "ram_gib": 64, "disk_gib": 0, "hour": 1.428, "month_cap": null }, { "name": "m7i.8xlarge", "vcpu": 32, "ram_gib": 128, "disk_gib": 0, "hour": 1.6968, "month_cap": null } ], "pool_tiers": null, "requires_always_on": false, "always_on_month_per_instance": null, "commit_note": null, "min_billed_seconds": 60, "granularity_s": 1, "start_fee": 0, "boot_overhead_s": null, "multiplier": 1, "windows_vcpu_h": 0.046, "gpu": null, "note": "ap-south-1 Mumbai c7i = us-east-1; ap-northeast-1 c7i.xlarge $0.2247, ap-southeast-1 $0.2058.", "product_class": "vm", "cpu_class": "dedicated" }, { "key": "on-demand-arm64", "label": "Graviton3 on-demand c7g/m7g (us-east-1)", "flags": [], "os": [ "linux" ], "regions": [ "us" ], "gpu_only": false, "features": { "arm64": true, "isolation": "vm", "nested_virt": false }, "pricing": "sizes", "vcpu_h": null, "ram_gib_h": null, "cpu_basis": "alloc", "ram_basis": "alloc", "active_floor": 0, "ram_per_vcpu": [ null, null ], "ram_fixed_per_vcpu": null, "vcpu_options": null, "min_vcpu": null, "max_vcpu": null, "max_ram_gib": null, "sizes": [ { "name": "c7g.large", "vcpu": 2, "ram_gib": 4, "disk_gib": 0, "hour": 0.0725, "month_cap": null }, { "name": "m7g.large", "vcpu": 2, "ram_gib": 8, "disk_gib": 0, "hour": 0.0816, "month_cap": null }, { "name": "c7g.xlarge", "vcpu": 4, "ram_gib": 8, "disk_gib": 0, "hour": 0.145, "month_cap": null }, { "name": "m7g.xlarge", "vcpu": 4, "ram_gib": 16, "disk_gib": 0, "hour": 0.1632, "month_cap": null }, { "name": "c7g.2xlarge", "vcpu": 8, "ram_gib": 16, "disk_gib": 0, "hour": 0.29, "month_cap": null }, { "name": "m7g.2xlarge", "vcpu": 8, "ram_gib": 32, "disk_gib": 0, "hour": 0.3264, "month_cap": null }, { "name": "c7g.4xlarge", "vcpu": 16, "ram_gib": 32, "disk_gib": 0, "hour": 0.58, "month_cap": null }, { "name": "m7g.4xlarge", "vcpu": 16, "ram_gib": 64, "disk_gib": 0, "hour": 0.6528, "month_cap": null }, { "name": "c7g.8xlarge", "vcpu": 32, "ram_gib": 64, "disk_gib": 0, "hour": 1.16, "month_cap": null }, { "name": "m7g.8xlarge", "vcpu": 32, "ram_gib": 128, "disk_gib": 0, "hour": 1.3056, "month_cap": null } ], "pool_tiers": null, "requires_always_on": false, "always_on_month_per_instance": null, "commit_note": null, "min_billed_seconds": 60, "granularity_s": 1, "start_fee": 0, "boot_overhead_s": null, "multiplier": 1, "windows_vcpu_h": null, "gpu": null, "note": "Graviton vCPU = full core. c8g.xlarge (Graviton4) $0.15952.", "product_class": "vm", "cpu_class": "dedicated" }, { "key": "on-demand-arm64-eu", "label": "Graviton3 on-demand (eu-north-1)", "flags": [], "os": [ "linux" ], "regions": [ "eu" ], "gpu_only": false, "features": { "arm64": true, "isolation": "vm", "nested_virt": false }, "pricing": "sizes", "vcpu_h": null, "ram_gib_h": null, "cpu_basis": "alloc", "ram_basis": "alloc", "active_floor": 0, "ram_per_vcpu": [ null, null ], "ram_fixed_per_vcpu": null, "vcpu_options": null, "min_vcpu": null, "max_vcpu": null, "max_ram_gib": null, "sizes": [ { "name": "c7g.large", "vcpu": 2, "ram_gib": 4, "disk_gib": 0, "hour": 0.0774, "month_cap": null }, { "name": "m7g.large", "vcpu": 2, "ram_gib": 8, "disk_gib": 0, "hour": 0.0867, "month_cap": null }, { "name": "c7g.xlarge", "vcpu": 4, "ram_gib": 8, "disk_gib": 0, "hour": 0.1547, "month_cap": null }, { "name": "m7g.xlarge", "vcpu": 4, "ram_gib": 16, "disk_gib": 0, "hour": 0.1734, "month_cap": null }, { "name": "c7g.2xlarge", "vcpu": 8, "ram_gib": 16, "disk_gib": 0, "hour": 0.3094, "month_cap": null }, { "name": "m7g.2xlarge", "vcpu": 8, "ram_gib": 32, "disk_gib": 0, "hour": 0.3468, "month_cap": null }, { "name": "c7g.4xlarge", "vcpu": 16, "ram_gib": 32, "disk_gib": 0, "hour": 0.6189, "month_cap": null }, { "name": "m7g.4xlarge", "vcpu": 16, "ram_gib": 64, "disk_gib": 0, "hour": 0.6936, "month_cap": null }, { "name": "c7g.8xlarge", "vcpu": 32, "ram_gib": 64, "disk_gib": 0, "hour": 1.2378, "month_cap": null }, { "name": "m7g.8xlarge", "vcpu": 32, "ram_gib": 128, "disk_gib": 0, "hour": 1.3872, "month_cap": null } ], "pool_tiers": null, "requires_always_on": false, "always_on_month_per_instance": null, "commit_note": null, "min_billed_seconds": 60, "granularity_s": 1, "start_fee": 0, "boot_overhead_s": null, "multiplier": 1, "windows_vcpu_h": null, "gpu": null, "note": "", "product_class": "vm", "cpu_class": "dedicated" }, { "key": "on-demand-arm64-asia", "label": "Graviton3 on-demand (ap-south-1, 32% below us-east-1)", "flags": [], "os": [ "linux" ], "regions": [ "asia" ], "gpu_only": false, "features": { "arm64": true, "isolation": "vm", "nested_virt": false }, "pricing": "sizes", "vcpu_h": null, "ram_gib_h": null, "cpu_basis": "alloc", "ram_basis": "alloc", "active_floor": 0, "ram_per_vcpu": [ null, null ], "ram_fixed_per_vcpu": null, "vcpu_options": null, "min_vcpu": null, "max_vcpu": null, "max_ram_gib": null, "sizes": [ { "name": "c7g.large", "vcpu": 2, "ram_gib": 4, "disk_gib": 0, "hour": 0.0491, "month_cap": null }, { "name": "m7g.large", "vcpu": 2, "ram_gib": 8, "disk_gib": 0, "hour": 0.0583, "month_cap": null }, { "name": "c7g.xlarge", "vcpu": 4, "ram_gib": 8, "disk_gib": 0, "hour": 0.0982, "month_cap": null }, { "name": "m7g.xlarge", "vcpu": 4, "ram_gib": 16, "disk_gib": 0, "hour": 0.1166, "month_cap": null }, { "name": "c7g.2xlarge", "vcpu": 8, "ram_gib": 16, "disk_gib": 0, "hour": 0.1963, "month_cap": null }, { "name": "m7g.2xlarge", "vcpu": 8, "ram_gib": 32, "disk_gib": 0, "hour": 0.2333, "month_cap": null }, { "name": "c7g.4xlarge", "vcpu": 16, "ram_gib": 32, "disk_gib": 0, "hour": 0.3926, "month_cap": null }, { "name": "m7g.4xlarge", "vcpu": 16, "ram_gib": 64, "disk_gib": 0, "hour": 0.4666, "month_cap": null }, { "name": "c7g.8xlarge", "vcpu": 32, "ram_gib": 64, "disk_gib": 0, "hour": 0.7853, "month_cap": null }, { "name": "m7g.8xlarge", "vcpu": 32, "ram_gib": 128, "disk_gib": 0, "hour": 0.9331, "month_cap": null } ], "pool_tiers": null, "requires_always_on": false, "always_on_month_per_instance": null, "commit_note": null, "min_billed_seconds": 60, "granularity_s": 1, "start_fee": 0, "boot_overhead_s": null, "multiplier": 1, "windows_vcpu_h": null, "gpu": null, "note": "", "product_class": "vm", "cpu_class": "dedicated" }, { "key": "burstable-t3", "label": "Burstable t3 (us-east-1, 30-40% baseline + $0.05/vCPU-h surplus)", "flags": [ "burstable" ], "os": [ "linux" ], "regions": [ "us" ], "gpu_only": false, "features": { "isolation": "vm", "nested_virt": false }, "pricing": "sizes", "vcpu_h": null, "ram_gib_h": null, "cpu_basis": "alloc", "ram_basis": "alloc", "active_floor": 0, "ram_per_vcpu": [ null, null ], "ram_fixed_per_vcpu": null, "vcpu_options": null, "min_vcpu": null, "max_vcpu": null, "max_ram_gib": null, "sizes": [ { "name": "t3.large", "vcpu": 2, "ram_gib": 8, "disk_gib": 0, "hour": 0.0832, "month_cap": null }, { "name": "t3.xlarge", "vcpu": 4, "ram_gib": 16, "disk_gib": 0, "hour": 0.1664, "month_cap": null }, { "name": "t3.2xlarge", "vcpu": 8, "ram_gib": 32, "disk_gib": 0, "hour": 0.3328, "month_cap": null } ], "pool_tiers": null, "requires_always_on": false, "always_on_month_per_instance": null, "commit_note": null, "min_billed_seconds": 60, "granularity_s": 1, "start_fee": 0, "boot_overhead_s": null, "multiplier": 1, "windows_vcpu_h": null, "gpu": null, "note": "Price List API. Unlimited-mode surplus CPU credits $0.05/vCPU-h Linux ($0.096 Windows, $0.04 t4g). Baseline per vCPU: t3.large 30%, t3.xlarge/2xlarge 40% (AWS docs credit table, verified 2026-09-28: t3.large 36 credits/h = 30%, t3.xlarge 96/h = 40%, t3.2xlarge 192/h = 40%; baseline_pct 40 = the xlarge/2xlarge sizes a 4-8 vCPU workload picks). t4g.xlarge (Graviton) $0.1344.", "baseline_pct": 40, "product_class": "vm", "cpu_class": "shared" }, { "key": "reserved-1y", "label": "1y RI / Instance SP no-upfront, x86 (us-east-1)", "flags": [ "commit" ], "os": [ "linux", "windows" ], "regions": [ "us" ], "gpu_only": false, "features": { "isolation": "vm", "nested_virt": false }, "pricing": "sizes", "vcpu_h": null, "ram_gib_h": null, "cpu_basis": "alloc", "ram_basis": "alloc", "active_floor": 0, "ram_per_vcpu": [ null, null ], "ram_fixed_per_vcpu": null, "vcpu_options": null, "min_vcpu": null, "max_vcpu": null, "max_ram_gib": null, "sizes": [ { "name": "c7i.large", "vcpu": 2, "ram_gib": 4, "disk_gib": 0, "hour": 0.05904, "month_cap": null }, { "name": "m7i.large", "vcpu": 2, "ram_gib": 8, "disk_gib": 0, "hour": 0.06668, "month_cap": null }, { "name": "c7i.xlarge", "vcpu": 4, "ram_gib": 8, "disk_gib": 0, "hour": 0.11808, "month_cap": null }, { "name": "m7i.xlarge", "vcpu": 4, "ram_gib": 16, "disk_gib": 0, "hour": 0.13336, "month_cap": null }, { "name": "c7i.2xlarge", "vcpu": 8, "ram_gib": 16, "disk_gib": 0, "hour": 0.23616, "month_cap": null }, { "name": "m7i.2xlarge", "vcpu": 8, "ram_gib": 32, "disk_gib": 0, "hour": 0.26672, "month_cap": null }, { "name": "c7i.4xlarge", "vcpu": 16, "ram_gib": 32, "disk_gib": 0, "hour": 0.47231, "month_cap": null }, { "name": "m7i.4xlarge", "vcpu": 16, "ram_gib": 64, "disk_gib": 0, "hour": 0.53343, "month_cap": null }, { "name": "c7i.8xlarge", "vcpu": 32, "ram_gib": 64, "disk_gib": 0, "hour": 0.94462, "month_cap": null }, { "name": "m7i.8xlarge", "vcpu": 32, "ram_gib": 128, "disk_gib": 0, "hour": 1.06687, "month_cap": null } ], "pool_tiers": null, "requires_always_on": true, "always_on_month_per_instance": null, "commit_note": "1-year Standard Reserved Instance, no upfront, us-east-1 (EC2 Instance Savings Plan rates are identical). Billed every hour of the term whether running or not.", "min_billed_seconds": 60, "granularity_s": 1, "start_fee": 0, "boot_overhead_s": null, "multiplier": 1, "windows_vcpu_h": 0.046, "gpu": null, "note": "34% off. Compute Savings Plan 1y no-upfront is costlier but flexible: c7i.xlarge $0.12899.", "commit_term": "1-year", "cpu_class": "dedicated" }, { "key": "reserved-3y", "label": "3y RI / Instance SP no-upfront, x86 (us-east-1)", "flags": [ "commit" ], "os": [ "linux", "windows" ], "regions": [ "us" ], "gpu_only": false, "features": { "isolation": "vm", "nested_virt": false }, "pricing": "sizes", "vcpu_h": null, "ram_gib_h": null, "cpu_basis": "alloc", "ram_basis": "alloc", "active_floor": 0, "ram_per_vcpu": [ null, null ], "ram_fixed_per_vcpu": null, "vcpu_options": null, "min_vcpu": null, "max_vcpu": null, "max_ram_gib": null, "sizes": [ { "name": "c7i.large", "vcpu": 2, "ram_gib": 4, "disk_gib": 0, "hour": 0.03929, "month_cap": null }, { "name": "m7i.large", "vcpu": 2, "ram_gib": 8, "disk_gib": 0, "hour": 0.04572, "month_cap": null }, { "name": "c7i.xlarge", "vcpu": 4, "ram_gib": 8, "disk_gib": 0, "hour": 0.07859, "month_cap": null }, { "name": "m7i.xlarge", "vcpu": 4, "ram_gib": 16, "disk_gib": 0, "hour": 0.09145, "month_cap": null }, { "name": "c7i.2xlarge", "vcpu": 8, "ram_gib": 16, "disk_gib": 0, "hour": 0.15717, "month_cap": null }, { "name": "m7i.2xlarge", "vcpu": 8, "ram_gib": 32, "disk_gib": 0, "hour": 0.18289, "month_cap": null }, { "name": "c7i.4xlarge", "vcpu": 16, "ram_gib": 32, "disk_gib": 0, "hour": 0.31435, "month_cap": null }, { "name": "m7i.4xlarge", "vcpu": 16, "ram_gib": 64, "disk_gib": 0, "hour": 0.36578, "month_cap": null }, { "name": "c7i.8xlarge", "vcpu": 32, "ram_gib": 64, "disk_gib": 0, "hour": 0.62869, "month_cap": null }, { "name": "m7i.8xlarge", "vcpu": 32, "ram_gib": 128, "disk_gib": 0, "hour": 0.73157, "month_cap": null } ], "pool_tiers": null, "requires_always_on": true, "always_on_month_per_instance": null, "commit_note": "3-year Standard Reserved Instance, no upfront, us-east-1 (EC2 Instance Savings Plan rates are identical). Billed every hour of the term whether running or not.", "min_billed_seconds": 60, "granularity_s": 1, "start_fee": 0, "boot_overhead_s": null, "multiplier": 1, "windows_vcpu_h": 0.046, "gpu": null, "note": "56% off. Compute SP 3y no-upfront c7i.xlarge $0.08618; 3y all-upfront Instance SP $0.0684.", "commit_term": "3-year", "cpu_class": "dedicated" }, { "key": "reserved-1y-arm64", "label": "1y RI no-upfront, Graviton (us-east-1)", "flags": [ "commit" ], "os": [ "linux" ], "regions": [ "us" ], "gpu_only": false, "features": { "arm64": true, "isolation": "vm", "nested_virt": false }, "pricing": "sizes", "vcpu_h": null, "ram_gib_h": null, "cpu_basis": "alloc", "ram_basis": "alloc", "active_floor": 0, "ram_per_vcpu": [ null, null ], "ram_fixed_per_vcpu": null, "vcpu_options": null, "min_vcpu": null, "max_vcpu": null, "max_ram_gib": null, "sizes": [ { "name": "c7g.large", "vcpu": 2, "ram_gib": 4, "disk_gib": 0, "hour": 0.0478, "month_cap": null }, { "name": "m7g.large", "vcpu": 2, "ram_gib": 8, "disk_gib": 0, "hour": 0.054, "month_cap": null }, { "name": "c7g.xlarge", "vcpu": 4, "ram_gib": 8, "disk_gib": 0, "hour": 0.0955, "month_cap": null }, { "name": "m7g.xlarge", "vcpu": 4, "ram_gib": 16, "disk_gib": 0, "hour": 0.108, "month_cap": null }, { "name": "c7g.2xlarge", "vcpu": 8, "ram_gib": 16, "disk_gib": 0, "hour": 0.1911, "month_cap": null }, { "name": "m7g.2xlarge", "vcpu": 8, "ram_gib": 32, "disk_gib": 0, "hour": 0.2159, "month_cap": null }, { "name": "c7g.4xlarge", "vcpu": 16, "ram_gib": 32, "disk_gib": 0, "hour": 0.3822, "month_cap": null }, { "name": "m7g.4xlarge", "vcpu": 16, "ram_gib": 64, "disk_gib": 0, "hour": 0.4318, "month_cap": null }, { "name": "c7g.8xlarge", "vcpu": 32, "ram_gib": 64, "disk_gib": 0, "hour": 0.7644, "month_cap": null }, { "name": "m7g.8xlarge", "vcpu": 32, "ram_gib": 128, "disk_gib": 0, "hour": 0.8637, "month_cap": null } ], "pool_tiers": null, "requires_always_on": true, "always_on_month_per_instance": null, "commit_note": "1-year Standard Reserved Instance, no upfront, us-east-1 (EC2 Instance Savings Plan rates are identical). Billed every hour of the term whether running or not.", "min_billed_seconds": 60, "granularity_s": 1, "start_fee": 0, "boot_overhead_s": null, "multiplier": 1, "windows_vcpu_h": null, "gpu": null, "note": "", "commit_term": "1-year", "cpu_class": "dedicated" }, { "key": "reserved-3y-arm64", "label": "3y RI no-upfront, Graviton (us-east-1)", "flags": [ "commit" ], "os": [ "linux" ], "regions": [ "us" ], "gpu_only": false, "features": { "arm64": true, "isolation": "vm", "nested_virt": false }, "pricing": "sizes", "vcpu_h": null, "ram_gib_h": null, "cpu_basis": "alloc", "ram_basis": "alloc", "active_floor": 0, "ram_per_vcpu": [ null, null ], "ram_fixed_per_vcpu": null, "vcpu_options": null, "min_vcpu": null, "max_vcpu": null, "max_ram_gib": null, "sizes": [ { "name": "c7g.large", "vcpu": 2, "ram_gib": 4, "disk_gib": 0, "hour": 0.0319, "month_cap": null }, { "name": "m7g.large", "vcpu": 2, "ram_gib": 8, "disk_gib": 0, "hour": 0.037, "month_cap": null }, { "name": "c7g.xlarge", "vcpu": 4, "ram_gib": 8, "disk_gib": 0, "hour": 0.0637, "month_cap": null }, { "name": "m7g.xlarge", "vcpu": 4, "ram_gib": 16, "disk_gib": 0, "hour": 0.074, "month_cap": null }, { "name": "c7g.2xlarge", "vcpu": 8, "ram_gib": 16, "disk_gib": 0, "hour": 0.1274, "month_cap": null }, { "name": "m7g.2xlarge", "vcpu": 8, "ram_gib": 32, "disk_gib": 0, "hour": 0.1481, "month_cap": null }, { "name": "c7g.4xlarge", "vcpu": 16, "ram_gib": 32, "disk_gib": 0, "hour": 0.2549, "month_cap": null }, { "name": "m7g.4xlarge", "vcpu": 16, "ram_gib": 64, "disk_gib": 0, "hour": 0.2961, "month_cap": null }, { "name": "c7g.8xlarge", "vcpu": 32, "ram_gib": 64, "disk_gib": 0, "hour": 0.5098, "month_cap": null }, { "name": "m7g.8xlarge", "vcpu": 32, "ram_gib": 128, "disk_gib": 0, "hour": 0.5922, "month_cap": null } ], "pool_tiers": null, "requires_always_on": true, "always_on_month_per_instance": null, "commit_note": "3-year Standard Reserved Instance, no upfront, us-east-1 (EC2 Instance Savings Plan rates are identical). Billed every hour of the term whether running or not.", "min_billed_seconds": 60, "granularity_s": 1, "start_fee": 0, "boot_overhead_s": null, "multiplier": 1, "windows_vcpu_h": null, "gpu": null, "note": "", "commit_term": "3-year", "cpu_class": "dedicated" }, { "key": "spot", "label": "Spot x86 (us-east-1, snapshot)", "flags": [ "spot" ], "os": [ "linux" ], "regions": [ "us" ], "gpu_only": false, "features": { "isolation": "vm", "nested_virt": false }, "pricing": "sizes", "vcpu_h": null, "ram_gib_h": null, "cpu_basis": "alloc", "ram_basis": "alloc", "active_floor": 0, "ram_per_vcpu": [ null, null ], "ram_fixed_per_vcpu": null, "vcpu_options": null, "min_vcpu": null, "max_vcpu": null, "max_ram_gib": null, "sizes": [ { "name": "c7i.large", "vcpu": 2, "ram_gib": 4, "disk_gib": 0, "hour": 0.0276, "month_cap": null }, { "name": "m7i.large", "vcpu": 2, "ram_gib": 8, "disk_gib": 0, "hour": 0.0417, "month_cap": null }, { "name": "c7i.xlarge", "vcpu": 4, "ram_gib": 8, "disk_gib": 0, "hour": 0.0491, "month_cap": null }, { "name": "m7i.xlarge", "vcpu": 4, "ram_gib": 16, "disk_gib": 0, "hour": 0.0814, "month_cap": null }, { "name": "c7i.2xlarge", "vcpu": 8, "ram_gib": 16, "disk_gib": 0, "hour": 0.1722, "month_cap": null }, { "name": "m7i.2xlarge", "vcpu": 8, "ram_gib": 32, "disk_gib": 0, "hour": 0.1675, "month_cap": null }, { "name": "c7i.4xlarge", "vcpu": 16, "ram_gib": 32, "disk_gib": 0, "hour": 0.2477, "month_cap": null }, { "name": "m7i.4xlarge", "vcpu": 16, "ram_gib": 64, "disk_gib": 0, "hour": 0.2937, "month_cap": null }, { "name": "c7i.8xlarge", "vcpu": 32, "ram_gib": 64, "disk_gib": 0, "hour": 0.4959, "month_cap": null }, { "name": "m7i.8xlarge", "vcpu": 32, "ram_gib": 128, "disk_gib": 0, "hour": 0.5522, "month_cap": null } ], "pool_tiers": null, "requires_always_on": false, "always_on_month_per_instance": null, "commit_note": null, "min_billed_seconds": 60, "granularity_s": 1, "start_fee": 0, "boot_overhead_s": null, "multiplier": 1, "windows_vcpu_h": null, "gpu": null, "note": "Point-in-time from https://website.spot.ec2.aws.a2z.com/spot.json (the data-url that aws.amazon.com/ec2/spot/pricing renders; lowest price per instance type in the region, updated every 5 min), fetched 2026-09-29 08:38 UTC. 2-minute interruption notice.", "cpu_class": "dedicated" }, { "key": "spot-eu", "label": "Spot x86 (eu-north-1, snapshot)", "flags": [ "spot" ], "os": [ "linux" ], "regions": [ "eu" ], "gpu_only": false, "features": { "isolation": "vm", "nested_virt": false }, "pricing": "sizes", "vcpu_h": null, "ram_gib_h": null, "cpu_basis": "alloc", "ram_basis": "alloc", "active_floor": 0, "ram_per_vcpu": [ null, null ], "ram_fixed_per_vcpu": null, "vcpu_options": null, "min_vcpu": null, "max_vcpu": null, "max_ram_gib": null, "sizes": [ { "name": "c7i.large", "vcpu": 2, "ram_gib": 4, "disk_gib": 0, "hour": 0.0155, "month_cap": null }, { "name": "m7i.large", "vcpu": 2, "ram_gib": 8, "disk_gib": 0, "hour": 0.0158, "month_cap": null }, { "name": "c7i.xlarge", "vcpu": 4, "ram_gib": 8, "disk_gib": 0, "hour": 0.035, "month_cap": null }, { "name": "m7i.xlarge", "vcpu": 4, "ram_gib": 16, "disk_gib": 0, "hour": 0.0332, "month_cap": null }, { "name": "c7i.2xlarge", "vcpu": 8, "ram_gib": 16, "disk_gib": 0, "hour": 0.0684, "month_cap": null }, { "name": "m7i.2xlarge", "vcpu": 8, "ram_gib": 32, "disk_gib": 0, "hour": 0.0793, "month_cap": null }, { "name": "c7i.4xlarge", "vcpu": 16, "ram_gib": 32, "disk_gib": 0, "hour": 0.1043, "month_cap": null }, { "name": "m7i.4xlarge", "vcpu": 16, "ram_gib": 64, "disk_gib": 0, "hour": 0.1054, "month_cap": null }, { "name": "c7i.8xlarge", "vcpu": 32, "ram_gib": 64, "disk_gib": 0, "hour": 0.2137, "month_cap": null }, { "name": "m7i.8xlarge", "vcpu": 32, "ram_gib": 128, "disk_gib": 0, "hour": 0.2556, "month_cap": null } ], "pool_tiers": null, "requires_always_on": false, "always_on_month_per_instance": null, "commit_note": null, "min_billed_seconds": 60, "granularity_s": 1, "start_fee": 0, "boot_overhead_s": null, "multiplier": 1, "windows_vcpu_h": null, "gpu": null, "note": "Point-in-time from https://website.spot.ec2.aws.a2z.com/spot.json (the data-url that aws.amazon.com/ec2/spot/pricing renders; lowest price per instance type in the region, updated every 5 min), fetched 2026-09-29 08:38 UTC. 2-minute interruption notice.", "cpu_class": "dedicated" }, { "key": "spot-asia", "label": "Spot x86 (ap-south-1, snapshot)", "flags": [ "spot" ], "os": [ "linux" ], "regions": [ "asia" ], "gpu_only": false, "features": { "isolation": "vm", "nested_virt": false }, "pricing": "sizes", "vcpu_h": null, "ram_gib_h": null, "cpu_basis": "alloc", "ram_basis": "alloc", "active_floor": 0, "ram_per_vcpu": [ null, null ], "ram_fixed_per_vcpu": null, "vcpu_options": null, "min_vcpu": null, "max_vcpu": null, "max_ram_gib": null, "sizes": [ { "name": "c7i.large", "vcpu": 2, "ram_gib": 4, "disk_gib": 0, "hour": 0.0399, "month_cap": null }, { "name": "m7i.large", "vcpu": 2, "ram_gib": 8, "disk_gib": 0, "hour": 0.0509, "month_cap": null }, { "name": "c7i.xlarge", "vcpu": 4, "ram_gib": 8, "disk_gib": 0, "hour": 0.0827, "month_cap": null }, { "name": "m7i.xlarge", "vcpu": 4, "ram_gib": 16, "disk_gib": 0, "hour": 0.0713, "month_cap": null }, { "name": "c7i.2xlarge", "vcpu": 8, "ram_gib": 16, "disk_gib": 0, "hour": 0.1538, "month_cap": null }, { "name": "m7i.2xlarge", "vcpu": 8, "ram_gib": 32, "disk_gib": 0, "hour": 0.1576, "month_cap": null }, { "name": "c7i.4xlarge", "vcpu": 16, "ram_gib": 32, "disk_gib": 0, "hour": 0.231, "month_cap": null }, { "name": "m7i.4xlarge", "vcpu": 16, "ram_gib": 64, "disk_gib": 0, "hour": 0.2776, "month_cap": null }, { "name": "c7i.8xlarge", "vcpu": 32, "ram_gib": 64, "disk_gib": 0, "hour": 0.4702, "month_cap": null }, { "name": "m7i.8xlarge", "vcpu": 32, "ram_gib": 128, "disk_gib": 0, "hour": 0.5768, "month_cap": null } ], "pool_tiers": null, "requires_always_on": false, "always_on_month_per_instance": null, "commit_note": null, "min_billed_seconds": 60, "granularity_s": 1, "start_fee": 0, "boot_overhead_s": null, "multiplier": 1, "windows_vcpu_h": null, "gpu": null, "note": "Point-in-time from https://website.spot.ec2.aws.a2z.com/spot.json (the data-url that aws.amazon.com/ec2/spot/pricing renders; lowest price per instance type in the region, updated every 5 min), fetched 2026-09-29 08:38 UTC. 2-minute interruption notice.", "cpu_class": "dedicated" }, { "key": "spot-arm64", "label": "Spot Graviton (us-east-1, snapshot)", "flags": [ "spot" ], "os": [ "linux" ], "regions": [ "us" ], "gpu_only": false, "features": { "arm64": true, "isolation": "vm", "nested_virt": false }, "pricing": "sizes", "vcpu_h": null, "ram_gib_h": null, "cpu_basis": "alloc", "ram_basis": "alloc", "active_floor": 0, "ram_per_vcpu": [ null, null ], "ram_fixed_per_vcpu": null, "vcpu_options": null, "min_vcpu": null, "max_vcpu": null, "max_ram_gib": null, "sizes": [ { "name": "c7g.large", "vcpu": 2, "ram_gib": 4, "disk_gib": 0, "hour": 0.0245, "month_cap": null }, { "name": "m7g.large", "vcpu": 2, "ram_gib": 8, "disk_gib": 0, "hour": 0.0386, "month_cap": null }, { "name": "c7g.xlarge", "vcpu": 4, "ram_gib": 8, "disk_gib": 0, "hour": 0.055, "month_cap": null }, { "name": "m7g.xlarge", "vcpu": 4, "ram_gib": 16, "disk_gib": 0, "hour": 0.0857, "month_cap": null }, { "name": "c7g.2xlarge", "vcpu": 8, "ram_gib": 16, "disk_gib": 0, "hour": 0.125, "month_cap": null }, { "name": "m7g.2xlarge", "vcpu": 8, "ram_gib": 32, "disk_gib": 0, "hour": 0.1449, "month_cap": null }, { "name": "c7g.4xlarge", "vcpu": 16, "ram_gib": 32, "disk_gib": 0, "hour": 0.2105, "month_cap": null }, { "name": "m7g.4xlarge", "vcpu": 16, "ram_gib": 64, "disk_gib": 0, "hour": 0.2386, "month_cap": null }, { "name": "c7g.8xlarge", "vcpu": 32, "ram_gib": 64, "disk_gib": 0, "hour": 0.4104, "month_cap": null }, { "name": "m7g.8xlarge", "vcpu": 32, "ram_gib": 128, "disk_gib": 0, "hour": 0.5501, "month_cap": null } ], "pool_tiers": null, "requires_always_on": false, "always_on_month_per_instance": null, "commit_note": null, "min_billed_seconds": 60, "granularity_s": 1, "start_fee": 0, "boot_overhead_s": null, "multiplier": 1, "windows_vcpu_h": null, "gpu": null, "note": "Point-in-time from https://website.spot.ec2.aws.a2z.com/spot.json (the data-url that aws.amazon.com/ec2/spot/pricing renders; lowest price per instance type in the region, updated every 5 min), fetched 2026-09-29 08:38 UTC. 2-minute interruption notice.", "cpu_class": "dedicated" }, { "key": "spot-arm64-eu", "label": "Spot Graviton (eu-north-1, snapshot)", "flags": [ "spot" ], "os": [ "linux" ], "regions": [ "eu" ], "gpu_only": false, "features": { "arm64": true, "isolation": "vm", "nested_virt": false }, "pricing": "sizes", "vcpu_h": null, "ram_gib_h": null, "cpu_basis": "alloc", "ram_basis": "alloc", "active_floor": 0, "ram_per_vcpu": [ null, null ], "ram_fixed_per_vcpu": null, "vcpu_options": null, "min_vcpu": null, "max_vcpu": null, "max_ram_gib": null, "sizes": [ { "name": "c7g.large", "vcpu": 2, "ram_gib": 4, "disk_gib": 0, "hour": 0.0196, "month_cap": null }, { "name": "m7g.large", "vcpu": 2, "ram_gib": 8, "disk_gib": 0, "hour": 0.0208, "month_cap": null }, { "name": "c7g.xlarge", "vcpu": 4, "ram_gib": 8, "disk_gib": 0, "hour": 0.0354, "month_cap": null }, { "name": "m7g.xlarge", "vcpu": 4, "ram_gib": 16, "disk_gib": 0, "hour": 0.035, "month_cap": null }, { "name": "c7g.2xlarge", "vcpu": 8, "ram_gib": 16, "disk_gib": 0, "hour": 0.0675, "month_cap": null }, { "name": "m7g.2xlarge", "vcpu": 8, "ram_gib": 32, "disk_gib": 0, "hour": 0.0598, "month_cap": null }, { "name": "c7g.4xlarge", "vcpu": 16, "ram_gib": 32, "disk_gib": 0, "hour": 0.0682, "month_cap": null }, { "name": "m7g.4xlarge", "vcpu": 16, "ram_gib": 64, "disk_gib": 0, "hour": 0.0762, "month_cap": null }, { "name": "c7g.8xlarge", "vcpu": 32, "ram_gib": 64, "disk_gib": 0, "hour": 0.1384, "month_cap": null }, { "name": "m7g.8xlarge", "vcpu": 32, "ram_gib": 128, "disk_gib": 0, "hour": 0.213, "month_cap": null } ], "pool_tiers": null, "requires_always_on": false, "always_on_month_per_instance": null, "commit_note": null, "min_billed_seconds": 60, "granularity_s": 1, "start_fee": 0, "boot_overhead_s": null, "multiplier": 1, "windows_vcpu_h": null, "gpu": null, "note": "Point-in-time from https://website.spot.ec2.aws.a2z.com/spot.json (the data-url that aws.amazon.com/ec2/spot/pricing renders; lowest price per instance type in the region, updated every 5 min), fetched 2026-09-29 08:38 UTC. 2-minute interruption notice.", "cpu_class": "dedicated" }, { "key": "spot-arm64-asia", "label": "Spot Graviton (ap-south-1, snapshot)", "flags": [ "spot" ], "os": [ "linux" ], "regions": [ "asia" ], "gpu_only": false, "features": { "arm64": true, "isolation": "vm", "nested_virt": false }, "pricing": "sizes", "vcpu_h": null, "ram_gib_h": null, "cpu_basis": "alloc", "ram_basis": "alloc", "active_floor": 0, "ram_per_vcpu": [ null, null ], "ram_fixed_per_vcpu": null, "vcpu_options": null, "min_vcpu": null, "max_vcpu": null, "max_ram_gib": null, "sizes": [ { "name": "c7g.large", "vcpu": 2, "ram_gib": 4, "disk_gib": 0, "hour": 0.0268, "month_cap": null }, { "name": "m7g.large", "vcpu": 2, "ram_gib": 8, "disk_gib": 0, "hour": 0.0326, "month_cap": null }, { "name": "c7g.xlarge", "vcpu": 4, "ram_gib": 8, "disk_gib": 0, "hour": 0.0551, "month_cap": null }, { "name": "m7g.xlarge", "vcpu": 4, "ram_gib": 16, "disk_gib": 0, "hour": 0.0589, "month_cap": null }, { "name": "c7g.2xlarge", "vcpu": 8, "ram_gib": 16, "disk_gib": 0, "hour": 0.0976, "month_cap": null }, { "name": "m7g.2xlarge", "vcpu": 8, "ram_gib": 32, "disk_gib": 0, "hour": 0.1047, "month_cap": null }, { "name": "c7g.4xlarge", "vcpu": 16, "ram_gib": 32, "disk_gib": 0, "hour": 0.1796, "month_cap": null }, { "name": "m7g.4xlarge", "vcpu": 16, "ram_gib": 64, "disk_gib": 0, "hour": 0.211, "month_cap": null }, { "name": "c7g.8xlarge", "vcpu": 32, "ram_gib": 64, "disk_gib": 0, "hour": 0.3354, "month_cap": null }, { "name": "m7g.8xlarge", "vcpu": 32, "ram_gib": 128, "disk_gib": 0, "hour": 0.4052, "month_cap": null } ], "pool_tiers": null, "requires_always_on": false, "always_on_month_per_instance": null, "commit_note": null, "min_billed_seconds": 60, "granularity_s": 1, "start_fee": 0, "boot_overhead_s": null, "multiplier": 1, "windows_vcpu_h": null, "gpu": null, "note": "Point-in-time from https://website.spot.ec2.aws.a2z.com/spot.json (the data-url that aws.amazon.com/ec2/spot/pricing renders; lowest price per instance type in the region, updated every 5 min), fetched 2026-09-29 08:38 UTC. 2-minute interruption notice.", "cpu_class": "dedicated" }, { "key": "on-demand-metal", "label": "Bare-metal on-demand c7i/m7i/c7g/m7g .metal (us-east-1)", "flags": [], "os": [ "linux" ], "regions": [ "us" ], "gpu_only": false, "features": { "nested_virt": true, "isolation": "bare-metal-vm" }, "pricing": "sizes", "vcpu_h": null, "ram_gib_h": null, "cpu_basis": "alloc", "ram_basis": "alloc", "active_floor": 0, "ram_per_vcpu": [ null, null ], "ram_fixed_per_vcpu": null, "vcpu_options": null, "min_vcpu": null, "max_vcpu": null, "max_ram_gib": null, "sizes": [ { "name": "c7g.metal (Arm)", "vcpu": 64, "ram_gib": 128, "disk_gib": 0, "hour": 2.32, "month_cap": null }, { "name": "m7g.metal (Arm)", "vcpu": 64, "ram_gib": 256, "disk_gib": 0, "hour": 2.6112, "month_cap": null }, { "name": "c7i.metal-24xl", "vcpu": 96, "ram_gib": 192, "disk_gib": 0, "hour": 4.284, "month_cap": null }, { "name": "m7i.metal-24xl", "vcpu": 96, "ram_gib": 384, "disk_gib": 0, "hour": 4.8384, "month_cap": null }, { "name": "c7i.metal-48xl", "vcpu": 192, "ram_gib": 384, "disk_gib": 0, "hour": 8.568, "month_cap": null }, { "name": "m7i.metal-48xl", "vcpu": 192, "ram_gib": 768, "disk_gib": 0, "hour": 9.6768, "month_cap": null } ], "pool_tiers": null, "requires_always_on": false, "always_on_month_per_instance": null, "commit_note": null, "min_billed_seconds": 60, "granularity_s": 1, "start_fee": 0, "boot_overhead_s": null, "multiplier": 1, "windows_vcpu_h": null, "gpu": null, "note": "Added 2026-09-28 (patches v5-calculators-aws-ec2 + v7-vps-aws-ec2): AWS Price List API us-east-1 Linux OnDemand RunInstances rows; .metal = whole physical host (KVM/nested virtualization possible). Exactly 24x/16x the xlarge prices. Arm and x86 sizes mixed in one list: pick arch explicitly. EBS/IPv4/egress extra. RIs for metal not modelled.", "cpu_class": "dedicated" }, { "key": "gpu-l4", "label": "GPU g6 (1x NVIDIA L4, us-east-1)", "flags": [], "os": [ "linux" ], "regions": [ "us" ], "gpu_only": true, "features": { "gpu": true, "isolation": "vm", "nested_virt": false }, "pricing": "sizes", "vcpu_h": null, "ram_gib_h": null, "cpu_basis": "alloc", "ram_basis": "alloc", "active_floor": 0, "ram_per_vcpu": [ null, null ], "ram_fixed_per_vcpu": null, "vcpu_options": null, "min_vcpu": null, "max_vcpu": null, "max_ram_gib": null, "sizes": [ { "name": "g6.xlarge", "vcpu": 4, "ram_gib": 16, "disk_gib": 0, "hour": 0, "month_cap": null }, { "name": "g6.2xlarge", "vcpu": 8, "ram_gib": 32, "disk_gib": 0, "hour": 0.1728, "month_cap": null }, { "name": "g6.4xlarge", "vcpu": 16, "ram_gib": 64, "disk_gib": 0, "hour": 0.5184, "month_cap": null }, { "name": "g6.8xlarge", "vcpu": 32, "ram_gib": 128, "disk_gib": 0, "hour": 1.2096, "month_cap": null }, { "name": "g6.16xlarge", "vcpu": 64, "ram_gib": 256, "disk_gib": 0, "hour": 2.592, "month_cap": null } ], "pool_tiers": null, "requires_always_on": false, "always_on_month_per_instance": null, "commit_note": null, "min_billed_seconds": 60, "granularity_s": 1, "start_fee": 0, "boot_overhead_s": null, "multiplier": 1, "windows_vcpu_h": null, "gpu": { "L4": 0.8048 }, "note": "g6.xlarge $0.8048/h. Source: AWS public pricing feed (b0.p.awsstatic.com meteredUnitMaps ec2-ondemand us-east-1 Linux), via patch v4-gpu-aws-ec2 2026-09-28. gpu = price of the smallest 1-GPU instance (its CPU/RAM included); size.hour = extra for larger 1-GPU hosts. Multi-GPU (4x/8x) instances not modelled.", "cpu_class": "dedicated" }, { "key": "gpu-a10g", "label": "GPU g5 (1x NVIDIA A10G, us-east-1)", "flags": [], "os": [ "linux" ], "regions": [ "us" ], "gpu_only": true, "features": { "gpu": true, "isolation": "vm", "nested_virt": false }, "pricing": "sizes", "vcpu_h": null, "ram_gib_h": null, "cpu_basis": "alloc", "ram_basis": "alloc", "active_floor": 0, "ram_per_vcpu": [ null, null ], "ram_fixed_per_vcpu": null, "vcpu_options": null, "min_vcpu": null, "max_vcpu": null, "max_ram_gib": null, "sizes": [ { "name": "g5.xlarge", "vcpu": 4, "ram_gib": 16, "disk_gib": 0, "hour": 0, "month_cap": null }, { "name": "g5.2xlarge", "vcpu": 8, "ram_gib": 32, "disk_gib": 0, "hour": 0.206, "month_cap": null }, { "name": "g5.4xlarge", "vcpu": 16, "ram_gib": 64, "disk_gib": 0, "hour": 0.618, "month_cap": null }, { "name": "g5.8xlarge", "vcpu": 32, "ram_gib": 128, "disk_gib": 0, "hour": 1.442, "month_cap": null }, { "name": "g5.16xlarge", "vcpu": 64, "ram_gib": 256, "disk_gib": 0, "hour": 3.09, "month_cap": null } ], "pool_tiers": null, "requires_always_on": false, "always_on_month_per_instance": null, "commit_note": null, "min_billed_seconds": 60, "granularity_s": 1, "start_fee": 0, "boot_overhead_s": null, "multiplier": 1, "windows_vcpu_h": null, "gpu": { "A10": 1.006 }, "note": "A10G listed under the engine name A10. g5.xlarge $1.006/h. Source: AWS public pricing feed (b0.p.awsstatic.com meteredUnitMaps ec2-ondemand us-east-1 Linux), via patch v4-gpu-aws-ec2 2026-09-28. gpu = price of the smallest 1-GPU instance (its CPU/RAM included); size.hour = extra for larger 1-GPU hosts. Multi-GPU (4x/8x) instances not modelled.", "cpu_class": "dedicated" }, { "key": "gpu-l40s", "label": "GPU g6e (1x NVIDIA L40S, us-east-1)", "flags": [], "os": [ "linux" ], "regions": [ "us" ], "gpu_only": true, "features": { "gpu": true, "isolation": "vm", "nested_virt": false }, "pricing": "sizes", "vcpu_h": null, "ram_gib_h": null, "cpu_basis": "alloc", "ram_basis": "alloc", "active_floor": 0, "ram_per_vcpu": [ null, null ], "ram_fixed_per_vcpu": null, "vcpu_options": null, "min_vcpu": null, "max_vcpu": null, "max_ram_gib": null, "sizes": [ { "name": "g6e.xlarge", "vcpu": 4, "ram_gib": 32, "disk_gib": 0, "hour": 0, "month_cap": null }, { "name": "g6e.2xlarge", "vcpu": 8, "ram_gib": 64, "disk_gib": 0, "hour": 0.38108, "month_cap": null }, { "name": "g6e.4xlarge", "vcpu": 16, "ram_gib": 128, "disk_gib": 0, "hour": 1.14324, "month_cap": null }, { "name": "g6e.8xlarge", "vcpu": 32, "ram_gib": 256, "disk_gib": 0, "hour": 2.66756, "month_cap": null }, { "name": "g6e.16xlarge", "vcpu": 64, "ram_gib": 512, "disk_gib": 0, "hour": 5.71619, "month_cap": null } ], "pool_tiers": null, "requires_always_on": false, "always_on_month_per_instance": null, "commit_note": null, "min_billed_seconds": 60, "granularity_s": 1, "start_fee": 0, "boot_overhead_s": null, "multiplier": 1, "windows_vcpu_h": null, "gpu": { "L40S": 1.861 }, "note": "g6e.xlarge $1.861/h. Source: AWS public pricing feed (b0.p.awsstatic.com meteredUnitMaps ec2-ondemand us-east-1 Linux), via patch v4-gpu-aws-ec2 2026-09-28. gpu = price of the smallest 1-GPU instance (its CPU/RAM included); size.hour = extra for larger 1-GPU hosts. Multi-GPU (4x/8x) instances not modelled.", "cpu_class": "dedicated" }, { "key": "gpu-h100", "label": "GPU p5.4xlarge (1x NVIDIA H100 SXM, us-east-1)", "flags": [], "os": [ "linux" ], "regions": [ "us" ], "gpu_only": true, "features": { "gpu": true, "isolation": "vm", "nested_virt": false }, "pricing": "sizes", "vcpu_h": null, "ram_gib_h": null, "cpu_basis": "alloc", "ram_basis": "alloc", "active_floor": 0, "ram_per_vcpu": [ null, null ], "ram_fixed_per_vcpu": null, "vcpu_options": null, "min_vcpu": null, "max_vcpu": null, "max_ram_gib": null, "sizes": [ { "name": "p5.4xlarge", "vcpu": 16, "ram_gib": 256, "disk_gib": 0, "hour": 0, "month_cap": null } ], "pool_tiers": null, "requires_always_on": false, "always_on_month_per_instance": null, "commit_note": null, "min_billed_seconds": 60, "granularity_s": 1, "start_fee": 0, "boot_overhead_s": null, "multiplier": 1, "windows_vcpu_h": null, "gpu": { "H100": 6.88 }, "note": "p5.4xlarge $6.88/h (= p5.48xlarge / 8). H200/B200/B300 are sold only as 8-GPU p5en/p6 instances and are not modelled. Source: AWS public pricing feed (b0.p.awsstatic.com meteredUnitMaps ec2-ondemand us-east-1 Linux), via patch v4-gpu-aws-ec2 2026-09-28. gpu = price of the smallest 1-GPU instance (its CPU/RAM included); size.hour = extra for larger 1-GPU hosts. Multi-GPU (4x/8x) instances not modelled.", "cpu_class": "dedicated" } ], "storage": { "included_disk_gib": 0, "disk_gib_month": 0.08, "disk_billed_when_stopped": true, "snapshot_gib_month": 0.05, "paused_ram_billed": false }, "network": { "egress_free_gib": 100, "egress_gib": 0.09, "ipv4_month": 3.65, "ipv4_available": true, "internet": { "default": "full" } }, "plans": [ { "name": "Pay-as-you-go (no platform fee)", "fee": 0, "flags": [], "fee_is_credit": false, "included_usd": 0, "per_seat": 0, "concurrency": null, "max_session_h": null, "max_vcpu": null, "max_ram_gib": null, "sizes_allowed": null, "note": "No platform fee; per-family vCPU quotas per account." } ], "free": { "monthly_credit": 0, "one_time_credit": 100 }, "features": { "linux": true, "windows": true, "macos": false, "arm64": true, "gpu": true, "gpu_types": [ "L4", "A10", "L40S", "H100" ], "snapshot": "fs", "pause_resume": true, "fork": false, "persistent_disk": true, "volumes": true, "auto_stop_idle": false, "max_session_h": null, "custom_image": true, "docker_inside": true, "nested_virt": true, "root": true, "systemd": true, "browser": false, "desktop": false, "computer_use": false, "code_interpreter": false, "agent_harnesses": false, "ssh": true, "public_ipv4": true, "inbound_https": false, "custom_domain": true, "raw_tcp_inbound": true, "egress_allowlist": true, "static_egress_ip": true, "private_network": true, "self_host": false, "byoc": false, "open_source": false, "regions": [ "us", "other", "eu", "asia" ], "soc2": true, "hipaa": true, "sso": true, "lang_python": true, "lang_node": false, "lang_go": null, "lang_rust": null, "lang_java": null, "gpu_desktop": false, "credential_injection": null, "egress_http_rules": true, "wake_on_request": false, "live_resize": false, "fork_running": false, "memory_snapshot_fork": false, "webhooks": null, "scheduled_wakeups": null, "audit_logs": null, "eu_data_residency": true, "zero_data_retention": null, "scoped_api_keys": null, "spend_limits": null, "mcp_server": false, "isolation": "vm", "image_oci": false, "snapshot_auto": true, "snapshot_on_demand": true, "harness_api": false, "egress_allowlist_level": "both", "ingress_rules": true, "guest_firewall": true, "secret_proxy": false, "volume_attach": true, "volume_shared": true, "anti_bot": false, "captcha_solving": false, "egress_ip_type": "static-datacenter", "own_agent_api": false, "max_session_unlimited": true, "open_internet": true, "usage_api": true }, "perf": { "cpu_runs_s": null, "cold_start_ms": null, "burst_tti_ms": null, "dax_s": null }, "knobs": [ { "id": "t3_cpu", "label": "t3 average CPU (surplus credits above baseline)", "kind": "select", "min": null, "max": null, "step": null, "default": 1, "unit": null, "options": [ { "value": 1, "label": "<= 40% avg CPU (within baseline)" }, { "value": 1.36, "label": "70% avg CPU (4 vCPU: +$0.06/h credits)" }, { "value": 1.72, "label": "100% avg CPU (4 vCPU: +$0.12/h credits)" } ], "effect": "rate_factor", "modes": [ "burstable-t3" ], "why": "t3 Unlimited bills $0.05 per vCPU-hour above the 40% baseline (t3.xlarge flat-out = $0.2864/h vs $0.1664 list)." }, { "id": "t3_surplus", "label": "t3 Unlimited: average CPU above baseline (share of each vCPU)", "kind": "slider", "min": 0, "max": 0.6, "step": 0.05, "default": 0, "unit": "%", "options": null, "effect": "rate_factor", "modes": [ "burstable-t3" ], "map": { "offset": 1, "scale": 1.202 }, "why": "T3 Unlimited bills surplus CPU credits at $0.05/vCPU-h (Linux) when the 24 h average exceeds the baseline (30% t3.large, 40% t3.xlarge/2xlarge). t3 on-demand is $0.0416/vCPU-h, so each +10 points above baseline adds ~12% (factor = 1 + 1.202 x excess). Source: aws.amazon.com/ec2/pricing/on-demand (T2/T3/T4g Unlimited Mode Pricing) and the EC2 burstable credit table." } ], "caveats": [ "On-demand and RI rates: AWS Price List API (AmazonEC2 index.csv, version 20260925174521) for us-east-1, eu-north-1, ap-south-1 (also checked eu-west-1, eu-central-1, ap-northeast-1, ap-southeast-1 for xlarge). \"Cheapest EU/Asia\" means cheapest among regions checked.", "Spot prices are a point-in-time snapshot of AWS's own spot feed and vary by AZ and hour. Savings Plan rates from AWSComputeSavingsPlan (2026-09-25).", "Reserved/SP modes are modelled only for us-east-1; regional RI rates (e.g. eu-north-1 c7i.xlarge 1y $0.12641, ap-south-1 Graviton 1y $0.065) are in the regime write-up.", "Windows licence-included = +$0.046/vCPU-h on c7i/m7i (on-demand and RI); t3 Windows is +$0.0184/vCPU-h; Graviton cannot run Windows; spot Windows spreads differ (c7i.xlarge spot Windows us-east-1 $0.2019, live spot.json 2026-09-29).", "Billing: per second with 60 s minimum (Linux, Windows). Stopped instances bill EBS and any public IPv4 but not compute.", "EBS gp3 $0.08/GB-month us-east-1 (eu-north-1 $0.0836, ap-south-1 $0.0912) incl. 3,000 IOPS/125 MB/s; snapshots $0.05/GB-month (eu-north-1 $0.0475). No free root disk: add ~8-80 GB gp3.", "Public IPv4: $0.005/h per address, in-use or idle (AmazonVPC price list, all regions checked) = $3.65/month at 730 h. Billed per ENI/instance that has a public IP.", "NAT gateway gotcha: tasks/functions in private subnets reach the internet through a NAT gateway at $0.045/h per gateway (~$32.85/month per AZ) plus $0.045/GB processed (us-east-1; eu-north-1 $0.046, ap-south-1 $0.056), charged ON TOP of the $0.09/GB egress. Not modelled; public IPv4 ($0.005/h) is the usual alternative.", "Internet egress (AWS Price List API, AWSDataTransfer us-east-1, 2026-09): first 100 GB/month free aggregated across all services and regions, then $0.09/GB first 10 TB, $0.085 next 40 TB, $0.07 next 100 TB, $0.05 over 150 TB. eu-west-1 is also $0.09; ap-south-1 $0.1093; ap-northeast-1 $0.114; ap-southeast-1 $0.12 (first-10-TB tier). The engine uses the flat us-east-1 first tier.", "Boot takes tens of seconds; not a fast sandbox. AWS Free Tier $200 credit for new accounts is generic.", "GPU modes cover single-GPU instances only (g6, g5, g6e, p5.4xlarge); 4x/8x bundles (H200/B200/B300, multi-L4/L40S) and fractional g6f slices are not modelled. Egress is priced at the first tier ($0.09/GB); AWS tiers are $0.09/$0.085/$0.07/$0.05 at 10/50/150 TB.", "t3 Unlimited surplus credits ($0.05/vCPU-h Linux above the 24 h average baseline) are not charged unless you set the t3 surplus knob; the default assumes average CPU at or below baseline. burstable-t3 is flagged burstable (40% baseline) so sustained-load workloads see it as a compromise.", "Reserved/Savings Plan regimes are flagged commit (1-year / 3-year); the RI is billed for every hour of the term whether or not the instance runs.", "Boot overhead not encoded (no measured cold start in research/benchmarks.json): EC2 instances take tens of seconds to boot and are billed from the running state (60 s minimum).", "Launch rate limits: RunInstances request bucket 5 (burst) refilled at 2/s, and a resource bucket of 1,000 instances refilled at 2 instances/s (StartInstances the same). A 1,000-instance burst is fine, after that ~7,200 launches/hour sustained; adjustable via Service Quotas.", "New accounts: Running On-Demand Standard (A,C,D,H,I,M,R,T,Z) quota default = 5 vCPUs per region (Spot Standard also 5; G/VT and P = 0). That is ONE 4-vCPU instance at a time until EC2 auto-raises the quota with usage or a Service Quotas request is approved; stopped instances don't count.", "Free plan (accounts created on/after 2025-07-15): $100 sign-up credit + up to $100 from activities, all expiring 12 months after sign-up. The Free plan (vs Paid plan) lasts 6 months or until credits run out, cannot overspend, excludes Savings Plans/RIs and services 'that would rapidly consume the entire AWS Free Tier credit amount', and the account is closed (data kept 90 days) unless upgraded. Free-tier-eligible EC2 types for these accounts are t3.micro/small, t4g.micro/small, c7i-flex.large, m7i-flex.large (whether other types are blocked on the Free plan is not stated verbatim on the pages checked). On the Paid plan the credits cover any EC2 usage. The legacy 750 h/month t2/t3.micro free tier only applies to accounts created before 2025-07-15.", "Cheaper 4 vCPU/8 GiB options the card does not model (us-east-1 Linux on-demand, Price List 20260925174521): c6g.xlarge (Graviton2) $0.136, t3a.xlarge (4/16, burstable) $0.1504, c6a.xlarge (AMD Milan) $0.153, c5a.xlarge $0.154, c8g.xlarge (Graviton4) $0.15952, c7i-flex.xlarge $0.16958 (Flex: full performance 'a majority of the time'), c6i.xlarge $0.17 vs modelled c7i.xlarge $0.1785 / c7g.xlarge $0.145. Newer c8i.xlarge $0.18742, c7a.xlarge $0.20528 are dearer.", "Cheapest Asia x86 region is Asia Pacific (Malaysia) ap-southeast-5 / (Thailand) ap-southeast-7: c7i.xlarge $0.17494 (-2% vs ap-south-1 $0.1785, ladder linear); ap-south-1 stays cheapest for m7i ($0.2121 vs $0.2142) and Graviton ($0.0982 vs $0.1416). EU: eu-north-1 confirmed cheapest of all 8 EU regions for c7i/m7i/c7g/m7g (eu-west-1 $0.19152, eu-south-2 $0.1915 c7i.xlarge).", "macOS is not in this card (see aws-ec2-mac): Mac instances run only on Dedicated Hosts with a 24-hour minimum allocation.", "T3/T4g launch in Unlimited mode by default (also for Spot), so the t3 surplus-credit charge applies unless the user switches to standard mode; new T8i burstable family surplus credits are $0.06/vCPU-h Linux, $0.115 Windows.", "Nested virtualization (/dev/kvm) on virtual EC2 instances exists only on C8i, M8i and R8i (whats-new, Feb 16, 2026); the modelled c7i/m7i modes cannot run KVM, so a nested-virt workload is priced on the .metal mode here. The cheaper real fit, c8i.xlarge at $0.18742/h us-east-1, is not modelled as a mode.", "Locations = the 34 commercial regions in the EC2 service-endpoints table (docs.aws.amazon.com/general/latest/gr/ec2-service.html, checked 2026-10-14), labels as AWS names them (e.g. Asia Pacific (Hyderabad) ap-south-2, (Malaysia) ap-southeast-5, (New Zealand) ap-southeast-6, (Taipei) ap-east-2). GovCloud (us-gov-east-1/us-gov-west-1, separate account) and China regions are excluded; AWS counts \"124 Availability Zones within 39 Geographic Regions\" in total." ], "sources": [ "https://aws.amazon.com/ec2/pricing/on-demand/", "https://pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonEC2/current/region_index.json", "https://pricing.us-east-1.amazonaws.com/savingsPlan/v1.0/aws/AWSComputeSavingsPlan/current/region_index.json", "https://website.spot.ec2.aws.a2z.com/spot.js", "https://pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonVPC/current/region_index.json", "https://pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AWSDataTransfer/current/region_index.json", "https://pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonEC2/current/us-east-1/index.json", "https://b0.p.awsstatic.com/pricing/2.0/meteredUnitMaps/ec2/USD/current/ec2-ondemand-without-sec-sel/US%20East%20(N.%20Virginia)/Linux/index.json", "https://aws.amazon.com/about-aws/whats-new/2026/02/amazon-ec2-nested-virtualization-on-virtual/", "https://website.spot.ec2.aws.a2z.com/spot.json", "https://docs.aws.amazon.com/general/latest/gr/ec2-service.html", "https://aws.amazon.com/about-aws/global-infrastructure/regions_az/" ], "fleet": { "default_max_instances": null, "default_max_vcpu": 5, "default_max_gpus": 0, "raise": "automatic", "raise_time_days": null, "verification": "Personal details, email and phone; payment card may be required.", "tiers": "EC2 grows On-Demand quotas with usage; requests via Service Quotas are also available.", "notes": "Published regional Standard-family vCPU baseline; applied quotas may differ for new accounts. G/VT and P GPU-family vCPU quotas start at zero. Windows shares EC2 quotas. Not an instance-count limit; capacity and IP quotas also constrain a fleet.", "sources": [ "https://docs.aws.amazon.com/ec2/latest/instancetypes/ec2-instance-quotas.html", "https://aws.amazon.com/ec2/faqs/", "https://aws.amazon.com/free/complete-signup/", "https://aws.amazon.com/free/complete-signup/" ], "researched": "2026-09-28" }, "audit_2026_09_29": { "applied": 35, "source": "checked against the provider's live pricing and docs pages", "reviewed": 9 }, "feature_evidence": { "snapshot_auto": { "value": true, "source": "https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/Hibernate.html", "quote": "Hibernation saves the contents from the instance memory (RAM) to your Amazon Elastic Block Store (Amazon EBS) root volume.", "note": "Only via hibernation, which must be enabled at launch (prerequisites: supported families/AMIs, encrypted root EBS big enough for RAM, RAM limits). Triggered by a hibernate-stop (StopInstances --hibernate) or automatically on Spot interruption if interruption behavior = hibernate ('You can specify that Amazon EC2 hibernates your Spot Instances when they are interrupted', interruption-behavior). On start: RAM reloaded, processes resumed. Memory included. A plain stop only keeps EBS (persistent_disk, not this). Scheduled EBS snapshots via Data Lifecycle Manager / AWS Backup exist ('AWS does not automatically back up the data stored on your EBS volumes').", "checked": "2026-09-29" }, "snapshot_on_demand": { "value": true, "source": "https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/creating-an-ami-ebs.html", "quote": "If you're confident that your instance is in a consistent state appropriate for AMI creation, you can tell Amazon EC2 not to power down and reboot the instance.", "note": "CreateImage (AMI) from a running or stopped instance (reboots by default for consistency, NoReboot optional) → launch any number of new instances from it (template/fork). EBS snapshots per volume via console/API ('You can create and manage your snapshots using the Amazon EC2 console or the Amazon EC2 API'), restorable to new volumes in any AZ of the region. Disk only; memory NOT included (only hibernation captures RAM, and it is not forkable).", "checked": "2026-09-29" }, "harness_api": { "value": false, "source": "https://aws.amazon.com/ec2/", "quote": "Amazon Elastic Compute Cloud (Amazon EC2) is a cloud computing service that provides resizable, secure compute capacity as virtual servers.", "note": "EC2 offers generic remote exec (SSM Run Command/Session Manager, user data) but no API to prompt/stream/follow-up with a coding agent harness (Claude Code, Codex, etc.) inside an instance. AWS's agent product for that is Bedrock AgentCore, a separate service not modelled on this card.", "checked": "2026-09-30" }, "egress_allowlist": { "value": true, "source": "https://docs.aws.amazon.com/vpc/latest/userguide/vpc-security-groups.html", "quote": "You can specify the destination, port range, and protocol for each outbound rule.", "note": "IP level: security group outbound rules (per instance/ENI, destination CIDR/prefix list/SG + port + protocol, free) and subnet NACLs; no-internet mode = private subnet without NAT/IGW. Domain level (paid add-ons, VPC-wide rather than per-instance unless routed per subnet): AWS Network Firewall stateful domain lists ('You can create allow lists and deny lists with domain names that the stateful rules engine looks for in network traffic'; SNI for HTTPS, Host header for HTTP) and Route 53 Resolver DNS Firewall ('you can deny access to all domains except for the ones that you explicitly trust'). All enforced outside the instance. Note: security groups do not filter traffic to Amazon DNS, DHCP, instance metadata, Time Sync.", "checked": "2026-09-29", "level": "both" }, "egress_http_rules": { "value": true, "source": "https://docs.aws.amazon.com/network-firewall/latest/developerguide/tls-inspection-configurations.html", "quote": "After decryption, Network Firewall inspects the traffic according to your firewall policy's stateful rules, and then re-encrypts it before sending it to its destination.", "note": "Via AWS Network Firewall (separately priced) Suricata-compatible stateful rules with HTTP keywords (examples use http.host, http.header, http.connection; Suricata also offers http.method/http.uri). For HTTPS this needs a TLS inspection configuration (firewall decrypts/re-encrypts with your CA). Not a per-instance EC2 setting.", "checked": "2026-09-29" }, "ingress_rules": { "value": true, "source": "https://docs.aws.amazon.com/vpc/latest/userguide/vpc-security-groups.html", "quote": "The security group acts as a virtual firewall. The only traffic that reaches the instance is the traffic allowed by the security group rules.", "note": "Security group inbound rules by source CIDR/prefix list/SG, port range, protocol; plus NACLs, private-only instances (no public IP), Network Firewall.", "checked": "2026-09-29" }, "guest_firewall": { "value": true, "source": null, "quote": null, "note": "Full Nitro/Xen VM with root and its own kernel; iptables/nftables/firewalld work as on any Linux VM.", "checked": "2026-09-29", "derived": true }, "secret_proxy": { "value": false, "source": "https://aws.amazon.com/ec2/", "quote": "Amazon Elastic Compute Cloud (Amazon EC2) is a cloud computing service that provides resizable, secure compute capacity as virtual servers.", "note": "No credential-injecting egress proxy exists for EC2. Secrets reach the guest as IAM instance-profile temporary credentials (instance metadata service) or via Secrets Manager/SSM Parameter Store calls made from inside the instance — a root user/process on the box can read them directly, unlike a proxy that keeps secrets off the machine.", "checked": "2026-09-30" }, "volume_attach": { "value": true, "source": "https://docs.aws.amazon.com/ebs/latest/userguide/ebs-attaching-volume.html", "quote": "You can attach an available EBS volume to one or more of your instances that is in the same Availability Zone as the volume.", "note": "EBS block volumes (gp3/gp2/io1/io2/st1/sc1), same-AZ, hot-attach/detach; max count depends on instance type. gp3 $0.08/GB-month us-east-1 (card).", "checked": "2026-09-29" }, "volume_shared": { "value": true, "source": "https://docs.aws.amazon.com/efs/latest/ug/how-it-works.html", "quote": "You can access your EFS file system concurrently from multiple NFS clients, so applications that scale beyond a single connection can access a file system.", "note": "Read-write. EFS (NFSv4.1, strong consistency and file locking, multi-AZ) and FSx family. Also EBS Multi-Attach for io1/io2 in one AZ: 'Each instance to which the volume is attached has full read and write permission to the shared volume' (needs a cluster-aware FS). Also Mountpoint for S3 (not checked here).", "checked": "2026-09-29" }, "anti_bot": { "value": false, "source": "https://aws.amazon.com/ec2/pricing/on-demand/", "quote": "Amazon EC2 provides secure, resizable compute capacity...to run virtually any workload", "note": "not offered: outside the product's scope — EC2 is a general-purpose VM, not a browser/scraping product with stealth/fingerprinting features", "checked": "2026-09-30" }, "captcha_solving": { "value": false, "source": "https://aws.amazon.com/ec2/pricing/on-demand/", "quote": "Amazon EC2 provides secure, resizable compute capacity...to run virtually any workload", "note": "not offered: outside the product's scope — EC2 is a general-purpose VM, no built-in CAPTCHA solving", "checked": "2026-09-30" }, "egress_ip_type": { "value": "static-datacenter", "source": "https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/elastic-ip-addresses-eip.html", "quote": "An Elastic IP address is a static IPv4 address designed for dynamic cloud computing. ... An Elastic IP address is a public IPv4 address, which is reachable from the internet.", "note": "AWS-owned datacenter IP ranges; Elastic IP gives a fixed (static) address, auto-assigned public IPv4 is also datacenter but not guaranteed static across stop/start unless an EIP is attached", "checked": "2026-09-30" }, "wake_on_request": { "value": false, "source": "https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/ec2-instance-resize.html", "quote": "You must stop your instance before you can change its instance type.", "note": "not offered: EC2 has no request-triggered auto-wake for stopped instances; instances must be started explicitly via StartInstances API/console/CLI/schedule, not by an incoming network request", "checked": "2026-09-30" }, "live_resize": { "value": false, "source": "https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/change-instance-type-of-ebs-backed-instance.html", "quote": "You must stop your instance before you can change its instance type. Ensure that you plan for downtime while your instance is stopped.", "note": "changing vCPU/RAM (instance type) requires stopping the instance first; not a live/running resize", "checked": "2026-09-30" }, "own_agent_api": { "value": false, "source": "https://aws.amazon.com/ec2/pricing/on-demand/", "quote": "Amazon EC2 provides secure, resizable compute capacity...to run virtually any workload", "note": "not offered: outside the product's scope — EC2 is raw compute; AWS's agent/browser-automation products (e.g. Bedrock AgentCore) are separate services, not part of EC2 itself", "checked": "2026-09-30" }, "mcp_server": { "value": false, "source": "https://github.com/awslabs/mcp", "quote": "Open source MCP Servers for AWS", "note": "awslabs/mcp ships general-purpose MCP servers for AWS APIs/IaC (e.g. AWS IaC MCP Server), not an MCP server dedicated to controlling EC2 sandboxes specifically", "checked": "2026-09-30" }, "lang_python": { "value": true, "source": "https://docs.aws.amazon.com/linux/al2023/ug/compare-with-al2.html", "quote": "Python 2.7 has been replaced with Python 3", "note": "Amazon Linux 2023, the default EC2 AMI family, ships Python 3 out of the box; other AMIs vary since EC2 lets you choose any OS image", "checked": "2026-09-30" }, "lang_node": { "value": false, "source": "https://docs.aws.amazon.com/linux/al2023/ug/compare-with-al2.html", "quote": "AL2023 contains thousands of software packages available for use.", "note": "Node.js is available as an installable package (dnf/yum or nvm) but is not preinstalled by default on Amazon Linux 2023 or other stock AMIs", "checked": "2026-09-30" }, "eu_data_residency": { "value": true, "source": "https://aws.amazon.com/about-aws/global-infrastructure/regions_az/", "quote": "The AWS Cloud spans 124 Availability Zones within 39 Geographic Regions", "note": "AWS operates multiple EU regions (e.g. eu-west-1 Ireland, eu-central-1 Frankfurt, eu-north-1 Stockholm) where instances and their data stay within the chosen region", "checked": "2026-09-30" }, "max_session_h": { "value": "none", "source": "https://aws.amazon.com/ec2/pricing/on-demand/", "quote": "Amazon EC2 provides secure, resizable compute capacity...pay only for what you use", "note": "on-demand EC2 instances run until you stop/terminate them; no maximum session duration is imposed (unlike e.g. Spot's interruption)", "checked": "2026-09-30" }, "fork": { "value": false, "source": "https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/creating-an-ami-ebs.html", "quote": "If you're confident that your instance is in a consistent state appropriate for AMI creation, you can tell Amazon EC2 not to power down and reboot the instance.", "note": "EC2 has no single-call live-clone/fork API. The only clone path is: create an AMI (disk-only, minutes, optionally without reboot) then separately RunInstances from it (a fresh boot, tens of seconds). Same taxonomy result as gcp-compute.json and azure-vm.json, which are also fork:false despite offering the identical AMI/image-launch pattern.", "checked": "2026-09-30" }, "fork_running": { "value": false, "source": "https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/creating-an-ami-ebs.html", "quote": "you can tell Amazon EC2 not to power down and reboot the instance", "note": "NoReboot AMI creation lets you snapshot disk state from a still-running instance, but the resulting clone is a brand-new instance booted from that AMI minutes later, not a live duplicate of the running machine kept in sync. No API clones a running instance in place. Consistent with gcp-compute/azure-vm being fork_running:false too.", "checked": "2026-09-30" }, "memory_snapshot_fork": { "value": false, "source": "https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/Hibernate.html", "quote": "Hibernation saves the contents from the instance memory (RAM) to your Amazon Elastic Block Store (Amazon EBS) root volume.", "note": "Hibernation persists RAM for resuming the SAME instance later (pause/resume), it does not clone that memory state into a NEW running instance. AMI/EBS snapshots (the only clone mechanism) are disk-only; memory is never included in a clone.", "checked": "2026-09-30" }, "desktop": { "value": false, "source": "https://docs.aws.amazon.com/AWSEC2/latest/WindowsGuide/EC2Win_GetStarted.html", "quote": "The following procedure uses the Remote Desktop Connection client for Windows (MSTSC).", "note": "reviewed: proposed correction rejected (This card's os list includes windows:true, and every Windows EC2 instance ships with RDP enabled by default, giving a full graphical desktop out of the box — the same reasoning already applied to give desktop:true in gcp-windows.json and azure-windows.json. Should be corrected to true (Windows only; Linux instances have no desktop unless you install one yourself).)", "checked": "2026-09-30" }, "browser": { "value": false, "source": "https://docs.aws.amazon.com/AWSEC2/latest/WindowsGuide/EC2Win_GetStarted.html", "quote": "The following procedure uses the Remote Desktop Connection client for Windows (MSTSC).", "note": "reviewed: proposed correction rejected (Windows Server AMIs (the OS this card prices) ship Microsoft Edge preinstalled, reachable once connected over the built-in RDP desktop — same basis used for browser:true on gcp-windows.json/azure-windows.json. Should be corrected to true for the Windows mode; Linux AMIs have no GUI browser by default.)", "checked": "2026-09-30" }, "computer_use": { "value": false, "source": "https://aws.amazon.com/ec2/", "quote": "Amazon Elastic Compute Cloud (Amazon EC2) is a cloud computing service that provides resizable, secure compute capacity as virtual servers.", "note": "EC2 is a generic VM product; it has no first-party screenshot+click computer-use API (that exists only as a separate AWS product, Bedrock AgentCore Browser/computer use, not modelled on this card).", "checked": "2026-09-30" }, "image_oci": { "value": false, "source": "https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/creating-an-ami-ebs.html", "quote": "You can create an Amazon EBS-backed Linux AMI from either a running or stopped instance.", "note": "Instances boot from AMIs (AWS's own machine-image format), never directly from a Docker/OCI image. Running an OCI image as the workload requires a container runtime installed inside the instance (docker_inside:true) or a different service (ECS/Fargate/EKS), not modelled here.", "checked": "2026-09-30" }, "auto_stop_idle": { "value": false, "source": "https://aws.amazon.com/ec2/faqs/", "quote": "Amazon EC2 provides three purchasing options... On-Demand Instances... you retain full control over its lifecycle", "note": "EC2 itself has no built-in auto-stop-on-idle setting; instances run until you (or your own CloudWatch alarm + Lambda/EventBridge automation, or a separate product like Cloud9) stop them. No first-party idle-shutdown feature on the EC2 API/console.", "checked": "2026-09-30" }, "inbound_https": { "value": false, "source": "https://docs.aws.amazon.com/vpc/latest/userguide/vpc-security-groups.html", "quote": "You can specify the destination, port range, and protocol for each outbound rule.", "note": "EC2 gives you a raw public IPv4/security-group ingress (raw_tcp_inbound:true) but no managed, ready-made HTTPS URL to an instance's port; TLS termination and a hostname require you to bring your own Route 53 domain, ACM certificate and either a load balancer or your own web-server config. There is no one-step 'exposed HTTPS URL' feature on the instance itself.", "checked": "2026-09-30" }, "agent_harnesses": { "value": false, "source": "https://aws.amazon.com/ec2/", "quote": "Amazon Elastic Compute Cloud (Amazon EC2) is a cloud computing service that provides resizable, secure compute capacity as virtual servers.", "note": "EC2 sells bare/generic AMIs (Amazon Linux, Ubuntu, Windows, etc.) and a Marketplace of third-party images; AWS does not preinstall or offer Claude Code / Codex / OpenCode as an official EC2 template.", "checked": "2026-09-30" }, "macos": { "value": false, "source": "https://aws.amazon.com/ec2/instance-types/mac/", "quote": "Amazon EC2 Mac Instances allow you to run on-demand macOS workloads in the cloud... Mac instances are available on Dedicated Hosts.", "note": "macOS EC2 instances exist, but only as Mac Dedicated Hosts with a mandatory minimum allocation period (24 hours), a completely different purchasing/billing model from the x86/Graviton on-demand and reserved modes priced on this card. This card's own caveats already say 'macOS is not in this card (see aws-ec2-mac)', so macos:false is correct for THIS card/product-scope; a true macOS offering exists under a separate card id.", "checked": "2026-09-30" }, "self_host": { "value": false, "source": "https://aws.amazon.com/ec2/", "quote": "Amazon Elastic Compute Cloud (Amazon EC2) is a cloud computing service that provides resizable, secure compute capacity as virtual servers.", "note": "EC2 is AWS's own hosted hypervisor/service (built on the proprietary Nitro system); there is no distributable EC2 software you can run on your own infrastructure outside AWS.", "checked": "2026-09-30" }, "byoc": { "value": false, "source": "https://aws.amazon.com/ec2/", "quote": "Amazon Elastic Compute Cloud (Amazon EC2) is a cloud computing service that provides resizable, secure compute capacity as virtual servers.", "note": "BYOC means a third-party managed service deployed into a customer's own cloud account; EC2 IS the cloud account's native compute primitive, not a third party's product deployed on top of someone else's cloud, so this concept does not apply.", "checked": "2026-09-30" }, "open_source": { "value": false, "source": "https://aws.amazon.com/ec2/", "quote": "Amazon Elastic Compute Cloud (Amazon EC2) is a cloud computing service that provides resizable, secure compute capacity as virtual servers.", "note": "EC2's virtualization stack (the proprietary Nitro hypervisor/system) and the EC2 service itself are closed-source AWS infrastructure; no OSI-licensed release of EC2 exists.", "checked": "2026-09-30" }, "cpu_class": { "value": { "on-demand": "dedicated", "on-demand-eu": "dedicated", "on-demand-asia": "dedicated", "on-demand-arm64": "dedicated", "on-demand-arm64-eu": "dedicated", "on-demand-arm64-asia": "dedicated", "burstable-t3": "burstable:40", "reserved-1y": "dedicated", "reserved-3y": "dedicated", "reserved-1y-arm64": "dedicated", "reserved-3y-arm64": "dedicated", "spot": "dedicated", "spot-eu": "dedicated", "spot-asia": "dedicated", "spot-arm64": "dedicated", "spot-arm64-eu": "dedicated", "spot-arm64-asia": "dedicated", "on-demand-metal": "dedicated", "gpu-l4": "dedicated", "gpu-a10g": "dedicated", "gpu-l40s": "dedicated", "gpu-h100": "dedicated" }, "source": "https://aws.amazon.com/ec2/faqs/", "quote": "If your workloads consistently require CPU usage much higher than the baseline, consider dedicated CPU instances such as the M or C.", "note": "M/C/T4g-non-burstable general purpose, compute-optimized, memory-optimized, GPU (P/G), and bare-metal instances give dedicated (non-shared) vCPUs; each vCPU is a full hardware thread/core reserved to the instance, not oversubscribed across tenants. Only the T-family (t3, t3a, t2, t4g) is burstable with a published baseline per vCPU. AWS's own instance-types page for T3 (https://aws.amazon.com/ec2/instance-types/t3/) publishes a 'Baseline Performance/vCPU' column: t3.large=30%, t3.xlarge=40%, t3.2xlarge=40% (t3.medium/small=20%, t3.micro=10%, t3.nano=5%). The card's 'burstable-t3' mode is keyed to the xlarge/2xlarge sizes (4-8 vCPU tier), so burstable:40 is used; t3.large alone would be burstable:30.", "checked": "2026-09-30" }, "usage_api": { "value": true, "source": "https://docs.aws.amazon.com/cost-management/latest/userguide/ce-api.html", "quote": "The Cost Explorer API lets you programmatically query your AWS costs and usage data. You can filter results by resource ID (e.g., EC2 instance ID) or by cost allocation tags. Use GetCostAndUsage operation to retrieve costs and metrics for specific instances or groups.", "note": "granularity: per-sandbox; cost: yes; AWS provides two primary APIs: Cost Explorer API (GetCostAndUsage) and Cost and Usage Report (CUR) export. Both support filtering by EC2 instance ID and custom tags, returning cost data in addition to usage quantities. CloudWatch provides metrics but those are monitoring data, not billable usage.", "confidence": "high", "checked": "2026-10-01" }, "nested_virt": { "value": true, "source": "https://aws.amazon.com/about-aws/whats-new/2026/02/amazon-ec2-nested-virtualization-on-virtual/", "quote": "This capability is available in all commercial regions on C8i, M8i, and R8i instances.", "note": "Card-level true comes from the on-demand-metal mode (.metal = whole host, KVM available). On virtual instances only C8i/M8i/R8i expose nested virtualization (posted Feb 16, 2026); the c7i/m7i modes (on-demand, reserved, spot) are set nested_virt:false per mode. c8i is not modelled as a mode (c8i.xlarge $0.18742 us-east-1, see caveats).", "checked": "2026-10-14" } }, "locations": [ { "country": "US", "state": "VA", "city": "N. Virginia", "provider_region": "us-east-1", "self_serve": true, "choosable": true }, { "country": "US", "state": "OH", "city": "Ohio", "provider_region": "us-east-2", "self_serve": true, "choosable": true }, { "country": "US", "state": "CA", "city": "N. California", "provider_region": "us-west-1", "self_serve": true, "choosable": true }, { "country": "US", "state": "OR", "city": "Oregon", "provider_region": "us-west-2", "self_serve": true, "choosable": true }, { "country": "CA", "state": null, "city": "Central", "provider_region": "ca-central-1", "self_serve": true, "choosable": true }, { "country": "CA", "state": null, "city": "West", "provider_region": "ca-west-1", "self_serve": true, "choosable": true }, { "country": "MX", "state": null, "city": "Mexico (Central)", "provider_region": "mx-central-1", "self_serve": true, "choosable": true }, { "country": "IE", "state": null, "city": "Ireland", "provider_region": "eu-west-1", "self_serve": true, "choosable": true }, { "country": "GB", "state": null, "city": "London", "provider_region": "eu-west-2", "self_serve": true, "choosable": true }, { "country": "FR", "state": null, "city": "Paris", "provider_region": "eu-west-3", "self_serve": true, "choosable": true }, { "country": "DE", "state": null, "city": "Frankfurt", "provider_region": "eu-central-1", "self_serve": true, "choosable": true }, { "country": "CH", "state": null, "city": "Zurich", "provider_region": "eu-central-2", "self_serve": true, "choosable": true }, { "country": "SE", "state": null, "city": "Stockholm", "provider_region": "eu-north-1", "self_serve": true, "choosable": true }, { "country": "IT", "state": null, "city": "Milan", "provider_region": "eu-south-1", "self_serve": true, "choosable": true }, { "country": "ES", "state": null, "city": "Spain", "provider_region": "eu-south-2", "self_serve": true, "choosable": true }, { "country": "ZA", "state": null, "city": "Cape Town", "provider_region": "af-south-1", "self_serve": true, "choosable": true }, { "country": "BH", "state": null, "city": "Bahrain", "provider_region": "me-south-1", "self_serve": true, "choosable": true }, { "country": "AE", "state": null, "city": "UAE", "provider_region": "me-central-1", "self_serve": true, "choosable": true }, { "country": "IL", "state": null, "city": "Tel Aviv", "provider_region": "il-central-1", "self_serve": true, "choosable": true }, { "country": "IN", "state": null, "city": "Mumbai", "provider_region": "ap-south-1", "self_serve": true, "choosable": true }, { "country": "IN", "state": null, "city": "Hyderabad", "provider_region": "ap-south-2", "self_serve": true, "choosable": true }, { "country": "JP", "state": null, "city": "Tokyo", "provider_region": "ap-northeast-1", "self_serve": true, "choosable": true }, { "country": "KR", "state": null, "city": "Seoul", "provider_region": "ap-northeast-2", "self_serve": true, "choosable": true }, { "country": "JP", "state": null, "city": "Osaka", "provider_region": "ap-northeast-3", "self_serve": true, "choosable": true }, { "country": "SG", "state": null, "city": "Singapore", "provider_region": "ap-southeast-1", "self_serve": true, "choosable": true }, { "country": "AU", "state": null, "city": "Sydney", "provider_region": "ap-southeast-2", "self_serve": true, "choosable": true }, { "country": "ID", "state": null, "city": "Jakarta", "provider_region": "ap-southeast-3", "self_serve": true, "choosable": true }, { "country": "AU", "state": null, "city": "Melbourne", "provider_region": "ap-southeast-4", "self_serve": true, "choosable": true }, { "country": "MY", "state": null, "city": "Malaysia", "provider_region": "ap-southeast-5", "self_serve": true, "choosable": true }, { "country": "NZ", "state": null, "city": "New Zealand", "provider_region": "ap-southeast-6", "self_serve": true, "choosable": true }, { "country": "TH", "state": null, "city": "Thailand", "provider_region": "ap-southeast-7", "self_serve": true, "choosable": true }, { "country": "HK", "state": null, "city": "Hong Kong", "provider_region": "ap-east-1", "self_serve": true, "choosable": true }, { "country": "TW", "state": null, "city": "Taipei", "provider_region": "ap-east-2", "self_serve": true, "choosable": true }, { "country": "BR", "state": null, "city": "São Paulo", "provider_region": "sa-east-1", "self_serve": true, "choosable": true } ], "locations_source": "https://docs.aws.amazon.com/general/latest/gr/ec2-service.html", "locations_quote": "If you specify the general endpoint (ec2.amazonaws.com), Amazon EC2 directs your request to the endpoint for `us-east-1`.", "locations_checked": "2026-10-14" }