m4.4xlarge pricing and specs in Asia Pacific (Mumbai)
16 vCPU · 64 GiB · High · Intel Xeon E5-2676 v3 (Haswell) previous generation
$0.8400/hour on-demand · $613/month (Linux, Shared)
m4.4xlarge is a general-purpose EC2 instance with 16 vCPU and 64 GiB, generation 4 of the m family; AWS lists it as a previous generation. On-demand it costs $0.8400 an hour, about $613 a month if it runs all month. A one-year commitment with nothing up front works out at $0.5235 an hour and pays for itself above roughly 454 hours a month; below that, on-demand is cheaper. The three-year options go down to $0.3081 an hour, 63% below on-demand. The newer m6g.4xlarge offers the same size for $0.4048 an hour, 52% less and moves to Graviton. Across regions the cheapest place to run m4.4xlarge is US East (N. Virginia) (us-east-1) at $0.8000 an hour, 5% below Asia Pacific (Mumbai).
Pricing by OS (hourly, USD, Shared)
| Option | Linux | Windows | RHEL | SUSE |
|---|---|---|---|---|
| On-Demand | $0.8400 | $1.5760 | $1.0128 | $0.9400 |
| Spot | $0.3088 | $0.8531 | — | — |
| 1y Std RI NU | $0.5235 | $1.2595 | $0.6963 | $0.5565 |
| 1y Std RI PU | $0.4986 | $1.2345 | $0.6714 | $0.5317 |
| 1y Std RI AU | $0.4886 | $1.2245 | $0.6614 | $0.5217 |
| 3y Std RI NU | $0.3540 | $1.0900 | $0.5268 | $0.3870 |
| 3y Std RI PU | $0.3278 | $1.0638 | $0.5006 | $0.3542 |
| 3y Std RI AU | $0.3081 | $1.0441 | $0.4809 | $0.3346 |
| 1y Conv RI NU | $0.6020 | $1.3380 | $0.7748 | $0.6350 |
| 3y Conv RI NU | $0.4071 | $1.1431 | $0.5799 | $0.4401 |
| 1y Compute SP NU | $0.6020 | $1.3380 | $0.7748 | $0.6350 |
| 3y Compute SP NU | $0.4071 | $1.1431 | $0.5799 | $0.4401 |
| 1y EC2 SP NU | $0.5235 | $1.2595 | $0.6963 | $0.5565 |
| 3y EC2 SP NU | $0.3540 | $1.0900 | $0.5268 | $0.3870 |
Spot by AZ (7-day p50, Linux): ap-south-1b $0.2978 · ap-south-1a $0.3197
Newer generations of the same size
Same family letter and size in a newer generation, ranked by price. “vCPU per $” uses vCPU as the performance proxy — newer chips usually add more, so it is a floor.
Graviton-eligible → m6g.4xlarge (-52%)| Type | $/h | Price Δ | vCPU per $ Δ | Saves / mo |
|---|---|---|---|---|
| m6g.4xlarge Graviton arch change | $0.4048 | -52% | +108% | $318 |
| m5a.4xlarge | $0.4440 | -47% | +89% | $289 |
| m6a.4xlarge | $0.4444 | -47% | +89% | $289 |
| m7g.4xlarge Graviton arch change | $0.4666 | -44% | +80% | $273 |
| m6gd.4xlarge Graviton arch change | $0.4832 | -42% | +74% | $260 |
| m8g.4xlarge Graviton arch change | $0.5133 | -39% | +64% | $239 |
| m5ad.4xlarge | $0.5370 | -36% | +56% | $221 |
| m7gd.4xlarge Graviton arch change | $0.5680 | -32% | +48% | $199 |
Cheapest regions for m4.4xlarge
| Region | On-Demand $/h | vs ap-south-1 | Saves / mo |
|---|---|---|---|
| US East (N. Virginia) us-east-1 | $0.8000 | -5% | $29 |
| US East (Ohio) us-east-2 | $0.8000 | -5% | $29 |
| US West (Oregon) us-west-2 | $0.8000 | -5% | $29 |
| Asia Pacific (Mumbai) ap-south-1 | $0.8400 | current | — |
| Canada (Central) ca-central-1 | $0.8880 | +6% | $-35 |
| EU (Ireland) eu-west-1 | $0.8880 | +6% | $-35 |
Caveat: instance price is only part of the bill. Cross-region and internet data transfer, latency to your users, data-residency rules and service availability can outweigh a few percent on compute.
Price history (ap-south-1)
Daily snapshots since the catalog started tracking this type; a row is added whenever a headline price changes.
| Date | OD Linux | OD Windows | 1y RI | 3y RI | Spot p50 |
|---|---|---|---|---|---|
| 2026-10-15 | $0.8400 | $1.5760 | $0.5235 | $0.3540 | $0.3088 |
| 2026-10-14 | $0.8400 | $1.5760 | $0.5235 | $0.3540 | $0.3101 |
| 2026-10-13 | $0.8400 | $1.5760 | $0.5235 | $0.3540 | $0.3114 |
| 2026-10-12 | $0.8400 | $1.5760 | $0.5235 | $0.3540 | $0.3137 |
| 2026-10-09 | $0.8400 | $1.5760 | $0.5235 | $0.3540 | $0.3153 |
| 2026-10-08 | $0.8400 | $1.5760 | $0.5235 | $0.3540 | $0.3151 |
| 2026-10-07 | $0.8400 | $1.5760 | $0.5235 | $0.3540 | $0.3133 |
| 2026-10-06 | $0.8400 | $1.5760 | $0.5235 | $0.3540 | $0.3131 |
| 2026-10-05 | $0.8400 | $1.5760 | $0.5235 | $0.3540 | $0.3129 |
| 2026-10-04 | $0.8400 | $1.5760 | $0.5235 | $0.3540 | $0.3125 |
| 2026-10-03 | $0.8400 | $1.5760 | $0.5235 | $0.3540 | $0.3124 |
| 2026-10-02 | $0.8400 | $1.5760 | $0.5235 | $0.3540 | $0.3000 |
| 2026-10-01 | $0.8400 | $1.5760 | $0.5235 | $0.3540 | $0.2848 |
| 2026-09-30 | $0.8400 | $1.5760 | $0.5235 | $0.3540 | $0.2725 |
| 2026-09-29 | $0.8400 | $1.5760 | $0.5235 | $0.3540 | $0.2627 |
| 2026-09-28 | $0.8400 | $1.5760 | $0.5235 | $0.3540 | $0.2575 |
| 2026-09-25 | $0.8400 | $1.5760 | $0.5235 | $0.3540 | $0.2551 |
| 2026-09-24 | $0.8400 | $1.5760 | $0.5235 | $0.3540 | $0.2555 |
| 2026-09-23 | $0.8400 | $1.5760 | $0.5235 | $0.3540 | $0.2583 |
| 2026-09-22 | $0.8400 | $1.5760 | $0.5235 | $0.3540 | $0.2612 |
All attributes
- vCPU
- 16
- Cores / threads per core
- 8 / 2
- Memory
- 64 GiB
- Memory per vCPU
- 4.00 GiB
- Processor
- Intel Xeon E5-2676 v3 (Haswell)
- Architecture
- x86_64
- Sustained clock
- 2.4 GHz
- Hypervisor
- xen
- Generation
- Previous
- Burstable
- No
- Bare metal
- No
- Free tier eligible
- No
- Network performance
- High
- Network baseline / burst
- 2.0 / 10.0 Gbps
- Max ENIs × IPv4 per ENI
- 8 × 30
- IPv6 / ENA / EFA
- IPv6 / — / —
- EBS optimized
- Yes
- EBS bandwidth (baseline / max)
- 2000 / 2000 Mbps
- EBS IOPS (baseline / max)
- 16,000 / 16000
- EBS NVMe
- unsupported
- Instance storage
- EBS only
- GPU
- —
- Placement groups
- cluster, partition, spread
- Hibernation / Nitro Enclaves
- Yes / No
- Dedicated hosts
- Supported
- Boot modes
- legacy-bios
- Spot supported
- Yes
Public AWS list prices (USD), refreshed nightly (catalog 2026-10-15). RI and Savings Plan rates are effective hourly (upfront amortised over the term) and are charged for every hour of the term. Excludes taxes, data transfer and storage. How these numbers are sourced and checked. Not affiliated with Amazon Web Services.
List prices are where it starts. Your bill is where it matters.
TruCost reads your actual AWS usage and tells you which of these commitments you should really buy, what your idle capacity costs, and where every dollar of the bill goes. Thirty minutes on your own data.