t4g.largevsm8gn.large
t4g.large is 54% cheaper than m8gn.large on demand in US East (N. Virginia), saving $57.16 a month.
| At a glance | t4g.large | m8gn.large |
|---|---|---|
| On-demand | $0.0672/hr54% less | $0.1455/hr |
| Lowest rate | $0.023/hr59% lessSpot, can be interrupted | $0.0565/hrSpot, can be interrupted |
| Size | 2 vCPUs, 8 GiB | 2 vCPUs, 8 GiB |
| Processor | AWS Graviton2 Processor | AWS Graviton4 Processor |
| Architecture | arm64 | arm64 |
| CPU | BurstableBest for workloads that are mostly idle | Full CPUSteady performance, no credits to manage |
Prices in US East (N. Virginia)
“Off” is the discount against each type’s own on-demand price. Linux, USD per hour.
| Option | t4g.large | m8gn.large | m8gn.large vs t4g.large |
|---|---|---|---|
| On-demandPay by the hour | $0.0672 | $0.1455 | ▲ 117% |
| SpotCan be interrupted | 66% off$0.023 | 61% off$0.0565 | ▲ 146% |
| 1-yr reservedNo upfront | 37% off$0.0421 | — | — |
| 3-yr reservedNo upfront | 57% off$0.029 | — | — |
| 1-yr savings planHourly commitment | 28% off$0.0485 | 27% off$0.1062 | ▲ 119% |
| 3-yr savings planHourly commitment | 50% off$0.0334 | 56% off$0.0644 | ▲ 93% |
Key differences
The biggest lead each one has.
Where t4g.large leads
On-demand price$0.0672 vs $0.1455 per hour
54% cheaperCPU modelEarns credits while idle, then bursts
BurstableIPv4 addresses per interface12 vs 10
20% higherWhere m8gn.large leads
CPU modelSteady performance, no credits to manage
Full CPUNetwork baseline6.25 Gbps vs 0.512 Gbps
12.2× higherEBS max bandwidth10,000 Mbps vs 2,780 Mbps
3.6× higherClock speed2.8 GHz vs 2.5 GHz
12% higherSpecs
19 of 39 specs differ. Those are in bold.
| Compute4 of 9 differ | t4g.large | m8gn.large |
|---|---|---|
| vCPUs | 2 | 2 |
| Physical cores | 2 | 2 |
| Threads per core | 1 | 1 |
| Architecture | arm64 | arm64 |
| Processor | AWS Graviton2 Processor | AWS Graviton4 Processor |
| CPU maker | AWS | AWS |
| Clock speed | 2.5 GHz | 2.8 GHz |
| Sustained clock | 2.5 GHz | 2.8 GHz |
| Burstable | Yes | No |
| MemoryAll 1 match | t4g.large | m8gn.large |
|---|---|---|
| Memory (GiB) | 8 | 8 |
| Networking4 of 8 differ | t4g.large | m8gn.large |
|---|---|---|
| Network | Up to 5 Gigabit | Up to 30 Gigabit |
| Network baseline | 0.512 Gbps | 6.25 Gbps |
| Network peak | 5 Gbps | 30 Gbps |
| Enhanced networking | Yes (ENA) | Yes (ENA) |
| EFA | No | No |
| IPv6 | Yes | Yes |
| Max network interfaces | 3 | 3 |
| IPv4 addresses per interface | 12 | 10 |
| Storage8 of 11 differ | t4g.large | m8gn.large |
|---|---|---|
| Storage | EBS only | EBS only |
| EBS optimized | Yes | Yes |
| EBS over NVMe | Required | Required |
| EBS throughput | Up to 2780 Mbps | Up to 10000 Mbps |
| EBS baseline bandwidth | 695 Mbps | 1,250 Mbps |
| EBS max bandwidth | 2,780 Mbps | 10,000 Mbps |
| EBS baseline throughput | 86.875 MB/s | 156.25 MB/s |
| EBS max throughput | 347.5 MB/s | 1,250 MB/s |
| EBS baseline IOPS | 4,000 | 5,000 |
| EBS max IOPS | 15,700 | 40,000 |
| Max EBS volumes | 27 | 32 |
| Instance details3 of 10 differ | t4g.large | m8gn.large |
|---|---|---|
| Family | t4g | m8gn |
| Category | General purpose | General purpose |
| Current generation | Yes | Yes |
| Normalization factor | 4 | 4 |
| Bare metal | No | No |
| Hypervisor | Nitro | Nitro |
| Placement groups | Partition, Spread | Cluster, Partition, Spread |
| Hibernation | Yes | Yes |
| Nitro Enclaves | No | Yes |
| Boot modes | UEFI | UEFI |
On-demand prices by region
Cheapest for t4g.large first. Linux, USD per hour.
| Region | t4g.large | m8gn.large | Difference |
|---|---|---|---|
| US East (N. Virginia)Used in summaryus-east-1 | $0.0672 | $0.1455 | ▲ 117% |
| US West (Oregon)us-west-2 | $0.0672 | $0.1455 | ▲ 117% |
| EU (Ireland)eu-west-1 | $0.0736 | $0.1626 | ▲ 121% |
Keep exploring
Each type’s own page, and how it compares with others.
AWS EC2 Linux prices in USD. Monthly figures use 730 hours. Prices last updated 12 Nov 2026.