M7G General Purpose
Balanced compute, memory, and networking for a wide range of workloads.
mGeneral Purposem = General PurposeBalanced compute, memory, and networking resources for a wide variety of standard enterprise workloads and microservices.77th Generation7 = 7th GenerationAWS 7th generation hardware platform with updated CPU/system architecture.gAWS Gravitong = AWS GravitonAWS Graviton3 Processor (arm64). Powered by custom 64-bit ARM AWS Graviton processors, delivering industry-leading price-performance and power efficiency.Alternatives to m7g
| Option | Price vs m7g | Why look at it |
|---|---|---|
| m7iCompare → | +24%+$0.0384/hr xlarge, Linux on-demand, us-east-1 | Intel (x86_64), same vCPU and RAM |
| m7aCompare → | +42%+$0.0686/hr xlarge, Linux on-demand, us-east-1 | AMD (x86_64), same vCPU and RAM |
| m7i-flexCompare → | +17%+$0.0283/hr xlarge, Linux on-demand, us-east-1 | Intel (x86_64), same vCPU and RAM |
| c7gCompare → | -11%-$0.0182/hr xlarge, Linux on-demand, us-east-1 | Different instance family, 8 GiB RAM instead of 16, same vCPU |
What to know about M7G General Purpose
Graviton3 general-purpose instances: the same 4 GiB per vCPU as m7i, at a lower hourly price.
The m7g is the Graviton3 counterpart to the m7i. It has the same 4 GiB per vCPU and similar networking and EBS limits, but it runs on AWS’s own Arm processors.
The draw is price. An m7g.large costs $0.0816 an hour in us-east-1 on Linux, about 19% less than the m7i.large. The gap holds across sizes because both families scale linearly.
The cost is compatibility. Container images need an arm64 build, and anything with native binaries, such as some language packages, agents, or vendor software, needs an arm64 version too. Test before you move a whole fleet. A pipeline that already builds multi-architecture images makes the switch much easier.
Sizes top out at 64 vCPUs, so very large single-node workloads may still need an m7i. This dataset lists only Linux prices for the m7g.
If you’re weighing it against the burstable t4g, the m7g is the choice for sustained load.
Good for
- Stateless services and containers built for arm64
- Replacing m7i or m6i when you can recompile or use multi-arch images
- Teams that want a lower bill without changing instance shape
Watch for
- Everything must run on arm64, including native dependencies
- Top size is 64 vCPUs, versus 192 on m7i
- Clock speed is 2.5 GHz, so compare benchmarks rather than GHz
Common use cases
- Web servers
- Application servers
- Small-to-medium databases
- Development environments
Head-to-Head Architecture Comparisons
View all comparisons →AWS M7 Generation: M7i vs M7a vs M7g vs M7i-flex
Comparing modern 4:1 memory-to-vCPU ratio General Purpose workhorses
Compare specifications →AWS General Purpose Evolution: M4 vs M5 vs M6i vs M7i vs M7g
From Haswell to Sapphire Rapids and Graviton3 in the 4:1 RAM workhorse tier
Compare specifications →RAM-to-vCPU Sizing Guide: C7g vs M7g vs R7g
Choosing between 2:1, 4:1, and 8:1 RAM-to-vCPU ratios on Graviton3
Compare specifications →General Purpose NVMe Comparison: M7g vs M7gd
Local high-speed NVMe storage for general purpose application tiers
Compare specifications →Instance types (8)
| Instance | vCPU | Memory | Architecture | From (Linux, $/hr) |
|---|---|---|---|---|
| m7g.medium | 1 | 4 GB | arm64 | $0.0292 |
| m7g.large | 2 | 8 GB | arm64 | $0.0583 |
| m7g.xlarge | 4 | 16 GB | arm64 | $0.1166 |
| m7g.2xlarge | 8 | 32 GB | arm64 | $0.2333 |
| m7g.4xlarge | 16 | 64 GB | arm64 | $0.4666 |
| m7g.8xlarge | 32 | 128 GB | arm64 | $0.9331 |
| m7g.12xlarge | 48 | 192 GB | arm64 | $1.3997 |
| m7g.16xlarge | 64 | 256 GB | arm64 | $1.8662 |