Quick answer
AWS documents two deployment choices from a SageMaker custom model’s details page:
- SageMaker AI Inference endpoints;
- Amazon Bedrock Custom Model Import.
Choose only after the model passes evaluation. The short deployment page establishes that both paths exist; it does not establish that every model, training method, artifact, or Region is compatible with both.
Decision checklist
| Decision | SageMaker endpoint | Bedrock import |
|---|---|---|
| Current model support | Verify SageMaker inference compatibility | Verify Bedrock Custom Model Import compatibility |
| API and client | Confirm endpoint contract and container behavior | Confirm Bedrock import and invocation contract |
| Capacity | Define serving configuration and scaling | Confirm supported capacity and quota model |
| Governance | Connect model package, lineage, IAM, and monitoring | Preserve source lineage and Bedrock governance controls |
| Region and data | Verify endpoint and artifact location | Verify import, storage, and inference Region |
| Cost | Model total endpoint and traffic cost | Model import and inference cost from live pricing |
This table is a review structure, not a product support promise. Resolve every row from current AWS documentation and your account.
Deployment gate
Before either path:
- compare the custom and base models on an immutable holdout;
- review safety, privacy, bias, security, and domain regressions;
- pin the model, dataset, job, evaluation, and artifact lineage;
- register the approved model version;
- define least-privilege invocation and administrative roles;
- test latency, throughput, errors, scaling, and cost;
- prepare canary, monitoring, rollback, and artifact retirement.
A registered model improves traceability; it does not prove deployment readiness. A successful import or endpoint creation does not prove output quality.
Avoid coupling training and serving decisions
Serverless customization manages training infrastructure. Serving is a separate lifecycle with its own availability, scaling, authentication, logging, retention, and billing. Do not describe the training job as “serverless inference” unless the selected serving product explicitly uses that term and configuration.
Use the data and evaluation guide before approval and the cost and operations guide for end-to-end economics.
Frequently asked questions
Where can a SageMaker customized open-weight model be deployed?
AWS documents two paths from the custom model details page: a SageMaker AI inference endpoint or Amazon Bedrock Custom Model Import.
Is every custom model deployable to both targets?
Do not assume so. Verify the exact model, artifact, technique, Region, quota, and current deployment documentation for each target before choosing it.
Should deployment start immediately after training completes?
No. Evaluate the logged model on an untouched holdout, review regressions and security, register the approved version, and define rollback and monitoring before serving traffic.
Official sources
- SageMaker open-weight model deployment
- SageMaker model customization job submission
- SageMaker model customization overview
Source check: August 4, 2026. Verify the exact model and artifact support, Regions, quotas, IAM, serving configuration, prices, monitoring, and import requirements before deployment.