Amazon SageMaker AI Instances Comparison

Batch Export API
Download complete pricing datasets

Discover and compare Amazon SageMaker machine learning instance specifications and pricing on a single page, covering various workload types and regions. Filter by GPU, inference accelerators, vCPUs, and memory. Compare prices across training, inference, and studio workloads.

The data is frequently updated from the AWS API, but may not be accurate. This site is not affiliated with Amazon or AWS. The latest update occurred on September 4, 2026 at 12:31 AM UTC

US Dollar ($)
Asia Pacific (Thailand) / ap-southeast-7
Per Hour
Real-Time Inference
On-Demand
vCPUs
to
Memory
to
Instance Type
Instance Family
vCPUs
Memory (GiB)
GPUs
Instance Price
Best Alternative Instances
Savings Options
Best Price Region / Diff
ml.c6g.12xlargeCompute Optimized Instances489601.9193-36%
ml.c6g.16xlargeCompute Optimized Instances6412802.5590-36%
ml.c6g.2xlargeCompute Optimized Instances81600.3199-36%
ml.c6g.4xlargeCompute Optimized Instances163200.6397-36%
ml.c6g.8xlargeCompute Optimized Instances326401.2794-36%
ml.c6g.largeCompute Optimized Instances2400.0799-36%
ml.c6g.xlargeCompute Optimized Instances4800.1600-36%
ml.c6gn.12xlargeCompute Optimized Instances489602.4238
ml.c6gn.16xlargeCompute Optimized Instances6412803.2317
ml.c6gn.2xlargeCompute Optimized Instances81600.4040
ml.c6gn.4xlargeCompute Optimized Instances163200.8079
ml.c6gn.8xlargeCompute Optimized Instances326401.6159
ml.c6gn.largeCompute Optimized Instances2400.1010
ml.c6gn.xlargeCompute Optimized Instances4800.2020
ml.c6i.12xlargeCompute Optimized Instances249602.3990
ml.c6i.16xlargeCompute Optimized Instances3212803.1987
ml.c6i.24xlargeCompute Optimized Instances4819204.7981
ml.c6i.2xlargeCompute Optimized Instances41600.3998
ml.c6i.32xlargeCompute Optimized Instances6425606.3974
ml.c6i.4xlargeCompute Optimized Instances83200.7997
ml.c6i.8xlargeCompute Optimized Instances166401.5994
ml.c6i.largeCompute Optimized Instances1400.1000
ml.c6i.xlargeCompute Optimized Instances2800.1999
ml.m6g.12xlargeStandard Instances4819202.3501-38%
ml.m6g.16xlargeStandard Instances6425603.1334-37%
ml.m6g.2xlargeStandard Instances83200.3917-37%
ml.m6g.4xlargeStandard Instances166400.7834-37%
ml.m6g.8xlargeStandard Instances3212801.5667-37%
ml.m6g.largeStandard Instances2800.0979-37%
ml.m6g.xlargeStandard Instances41600.1958-38%
ml.m6gd.12xlargeStandard Instances4819202.7660-37%
ml.m6gd.16xlargeStandard Instances6425603.6880-37%
ml.m6gd.2xlargeStandard Instances83200.4610-37%
ml.m6gd.4xlargeStandard Instances166400.9220-37%
ml.m6gd.8xlargeStandard Instances3212801.8439-37%
ml.m6gd.largeStandard Instances2800.1152-37%
ml.m6gd.xlargeStandard Instances41600.2305-37%
ml.m6i.12xlargeStandard Instances2419202.9376-5%
ml.m6i.16xlargeStandard Instances3225603.9168-5%
ml.m6i.24xlargeStandard Instances4838405.8752-5%
ml.m6i.2xlargeStandard Instances43200.4896-6%
ml.m6i.32xlargeStandard Instances6451207.8336-5%
ml.m6i.4xlargeStandard Instances86400.9792-5%
ml.m6i.8xlargeStandard Instances1612801.9584-5%
ml.m6i.largeStandard Instances1800.1224-6%
ml.m6i.xlargeStandard Instances21600.2448-6%
ml.r6g.12xlargeStandard Instances4838402.9767-37%
ml.r6g.16xlargeStandard Instances6451203.9690-37%
ml.r6g.2xlargeStandard Instances86400.4961-37%
ml.r6g.4xlargeStandard Instances1612800.9923-37%
ml.r6g.8xlargeStandard Instances3225601.9846-37%
ml.r6g.largeStandard Instances21600.1241-37%
ml.r6g.xlargeStandard Instances43200.2480-37%
ml.r6i.12xlargeMemory Optimized Instances2438403.7210-2%
ml.r6i.16xlargeMemory Optimized Instances3251204.9613-2%
ml.r6i.24xlargeMemory Optimized Instances4876807.4419-2%
ml.r6i.2xlargeMemory Optimized Instances46400.6202-2%
ml.r6i.32xlargeMemory Optimized Instances64102409.9226-2%
ml.r6i.4xlargeMemory Optimized Instances812801.2403-2%
ml.r6i.8xlargeMemory Optimized Instances1625602.4806-2%
ml.r6i.largeMemory Optimized Instances11600.1550-2%
ml.r6i.xlargeMemory Optimized Instances23200.3101-2%
ml.c7g.12xlargeCompute Optimized Instances489602.0390-30%
ml.c7g.16xlargeCompute Optimized Instances6412802.7187-30%
ml.c7g.2xlargeCompute Optimized Instances81600.3398-30%
ml.c7g.4xlargeCompute Optimized Instances163200.6797-30%
ml.c7g.8xlargeCompute Optimized Instances326401.3594-30%
ml.c7g.largeCompute Optimized Instances2400.0850-30%
ml.c7g.xlargeCompute Optimized Instances4800.1699-30%
ml.c7i.12xlargeCompute Optimized Instances249602.5191
ml.c7i.16xlargeCompute Optimized Instances3212803.3588
ml.c7i.24xlargeCompute Optimized Instances4819205.0383
ml.c7i.2xlargeCompute Optimized Instances41600.4199
ml.c7i.48xlargeCompute Optimized Instances96384010.0765
ml.c7i.4xlargeCompute Optimized Instances83200.8397
ml.c7i.8xlargeCompute Optimized Instances166401.6794
ml.c7i.largeCompute Optimized Instances1400.1050
ml.c7i.xlargeCompute Optimized Instances2800.2099
ml.m7i.12xlargeStandard Instances2419203.0845-5%
ml.m7i.16xlargeStandard Instances3225604.1126-5%
ml.m7i.24xlargeStandard Instances4838406.1690-5%
ml.m7i.2xlargeStandard Instances43200.5141-5%
ml.m7i.48xlargeStandard Instances96768012.3379-5%
ml.m7i.4xlargeStandard Instances86401.0282-5%
ml.m7i.8xlargeStandard Instances1612802.0563-5%
ml.m7i.largeStandard Instances1800.1285-5%
ml.m7i.xlargeStandard Instances21600.2570-5%
ml.r7i.12xlargeMemory Optimized Instances2438403.9070-2%
ml.r7i.16xlargeMemory Optimized Instances3251205.2093-2%
ml.r7i.24xlargeMemory Optimized Instances4876807.8140-2%
ml.r7i.2xlargeMemory Optimized Instances46400.6512-2%
ml.r7i.48xlargeMemory Optimized Instances961536015.6280-2%
ml.r7i.4xlargeMemory Optimized Instances812801.3023-2%
ml.r7i.8xlargeMemory Optimized Instances1625602.6047-2%
ml.r7i.largeMemory Optimized Instances11600.1628-2%
ml.r7i.xlargeMemory Optimized Instances23200.3256-2%