Amazon Redshift MLvs
AWS SageMaker
AWS SageMaker: Build, train, and deploy machine learning models at scale
Amazon Redshift ML
SQL statements in Redshift that train models on SageMaker and return them as functions
As of 30 August 2026, Amazon Redshift ML is free to use. Redshift ML lets an analyst create and use a model with SQL, with training handed to SageMaker behind the scenes. Softwr lists it under Machine Learning. Amazon Redshift ML is made by Amazon Web Services, launched in 2006, available on Web.
Overview
What Amazon Redshift ML does
Redshift ML is a feature of Amazon Redshift rather than a product of its own. A CREATE MODEL statement exports the selected rows to S3, passes them to Amazon SageMaker, in the default path SageMaker Autopilot, which trains and evaluates a set of candidate models and selects one, and then compiles the winner back into the Redshift cluster as a SQL function you call inside a SELECT. It also supports bringing an existing model, either compiled locally into the cluster or invoked remotely against a SageMaker endpoint. The analyst never opens a notebook, never configures training infrastructure and never leaves SQL. What it removes is the handoff. In most organisations the data is in the warehouse and the modelling happens somewhere else, which means an export, a schedule, a second set of credentials and a standing argument about who owns the resulting pipeline. Redshift ML collapses that into a statement someone who already has warehouse access can write this afternoon. That is worth a great deal to an analytics team whose alternative is a six-week queue in front of a small data science group, and it is worth close to nothing to a team that already has a functioning ML platform. It is used by analytics teams inside Redshift shops who want a propensity, churn or forecast column available to a dashboard. The trade-off is two invoices and one cloud. Training runs on SageMaker and appears on the SageMaker line, where the analyst who ran the statement usually cannot see it, and Autopilot will explore many candidates unless the data size or the MAX_CELLS setting constrains it. The resulting model is also inseparable from AWS: local inference executes inside the Redshift cluster and the artefact is not something you take to another warehouse.
What people use it for
- Adding a churn or propensity score to an existing dashboard where the data is already in Redshift and nobody needs a bespoke model
- Letting an analytics team test whether a predictive column has any business value before asking for data science headcount
- Scoring rows inside a SQL pipeline where moving data out to a separate service would add fragility for little benefit
- Organisations committed to AWS whose main constraint is a data science backlog rather than modelling sophistication
Related searches
The honest half
Where it falls short
Concrete and checkable, so you can decide whether any of them matter to you. This is the half of a review a vendor will not write about Amazon Redshift ML.
- Training is billed by SageMaker separately from Redshift, so a feature that looks like a free SQL statement produces a second line item on a different part of the bill that the analyst who ran it usually cannot see.
- Autopilot searches many candidate models by default and the duration and cost of CREATE MODEL scale with the data size and the MAX_CELLS setting, so an unconstrained statement against a large table is an expensive accident rather than an experiment.
- Local inference runs on the Redshift cluster itself, so scoring millions of rows competes for the resources the warehouse exists to provide, and the remote inference alternative adds a per-batch network call plus an hourly SageMaker endpoint charge that persists whether or not anyone queries it.
- The supported problem types are limited to what the exposed algorithms cover, so anything involving text, images, sequences, a custom loss function or a bespoke evaluation metric is out of scope and has to be built conventionally.
- There is no retraining schedule, drift detection or model registry, so a model created by a statement stays exactly as trained until somebody remembers to recreate it, and nothing in the warehouse will report that its accuracy has decayed.
Cross-shopped
What people choose instead of Amazon Redshift ML
Each pairing was judged by two reviewers asking whether a buyer would genuinely weigh the two against each other. The ones that failed were deleted rather than published.
- Amazon Redshift MLvsGoogle Vertex AI
Google Vertex AI: Unified ML platform to build, deploy, and scale AI models
- Amazon Redshift MLvsAzure Machine Learning
Azure Machine Learning: Enterprise-grade machine learning service
- Amazon Redshift MLvsDataRobot
DataRobot: Enterprise AI platform for automated machine learning
Pricing
What Amazon Redshift ML costs
Taken from the vendor's own pricing page. Prices move, so check before you buy.
Free Trial
Free
- 2-month trial
- 750 DC2.Large hours
On-Demand
$0.25 /hour
- Per-node pricing
- SageMaker training
Capabilities
Features
CREATE MODEL in SQL
Trains a model from a SELECT statement without leaving the warehouse interface
Automatic model selection
SageMaker Autopilot explores candidate algorithms and hyperparameters and returns the best performer
Local inference
The chosen model is compiled into the Redshift cluster so predictions run as a SQL function with no network call
Bring your own model
Points at an existing SageMaker-trained model, either compiled locally or invoked as a remote endpoint
Algorithm selection
Supports specifying XGBoost, linear learner, multilayer perceptron or K-means directly instead of automatic search
Cost ceiling controls
MAX_CELLS and related settings bound how much data and search the training job will consume
Existing warehouse security
Model access follows the same Redshift grants and IAM roles as the underlying tables
Batch and interactive scoring
Predictions run inline in any query, so scores can be joined and filtered like any other column
Answered, with sources
Questions people ask
Each answer names the page it came from, so you can check it rather than take our word for it.
Does it require SageMaker?
Yes. Redshift ML is an interface; the training happens in SageMaker and needs an IAM role and an S3 bucket for the intermediate data.
Is there an extra charge?
The SQL interface is part of Redshift, but the training runs as a SageMaker job charged at SageMaker rates, and a remote inference endpoint is billed for as long as it exists.
What kinds of model can it build?
Regression, binary and multiclass classification through the automatic path, plus direct use of XGBoost, linear learner, multilayer perceptron and K-means. Anything beyond structured tabular prediction is out of scope.
Can I use a model I trained myself?
Yes, through the bring-your-own-model path, either compiled into the cluster for local inference or called as a remote SageMaker endpoint.
Does it retrain automatically?
No. Retraining means running CREATE MODEL again, on a schedule you build yourself, and nothing in the product monitors whether it is needed.
Behind it
Who makes Amazon Redshift ML
- Company
- Amazon Web Services
- Based in
- Seattle, Washington
Keep looking
Where to go from Amazon Redshift ML
- Amazon Redshift ML alternativesEvery product people weigh against it
- Amazon Redshift ML pricingPlans, floors and what the tiers exclude
- Amazon Redshift ML badgesWhich rules it meets, and the figure behind each
- All Machine Learning softwareThe full category
- Best Machine Learning softwareThe shortlist, and how it was cut
Best Machine Learning software for
Compare Amazon Redshift ML with
- Google Vertex AI
- DataRobot
- Azure Machine Learning
- MATLAB
- Apache Spark MLlib
- RapidMiner
- Databricks
- SAS
- scikit-learn
- H2O.ai
- Weka
- DVC
- Hugging Face
- Kubeflow
- Langwatch
- LlamaIndex
- BigQuery ML
- H2O.ai
Other Machine Learning software
Microsoft's managed platform for training, tracking and deploying models on Azure
Free plan13 researched notesThe machine learning library inside Apache Spark, for data that will not fit on one machine
Free plan14 researched notesVisual workflow data science platform, now sold by Altair as AI Studio
Free plan14 researched notesUnified analytics platform for data engineering and data science
Free plan13 researched notesGit-style versioning for data sets and models, with the files kept in object storage
Free plan14 researched notesLLM engineering platform for testing and evaluating AI agents in production
Free, then €29/mo10 researched notes
Softwr does not host reviews and shows no star rating for Amazon Redshift ML, because a rating we did not collect is not ours to publish. What is here is the pricing and platform detail from the vendor’s own pages, limitations we could state concretely, and alternatives a reviewer confirmed people weigh against it. Tell us if any of it is wrong.
More on Amazon Redshift ML
Amazon Redshift ML alternatives
What people switch to, and what they give up
Amazon Redshift ML pricing
Every tier, and where the cost actually lands
Compare Amazon Redshift ML
Put it head to head with anything we hold
Amazon Redshift ML badge
Its rating, and an embed for your own site
Best Machine Learning software alternatives
Open-source MLOps platform for experiment tracking and orchestration
Open-source self-hosted, with paid hosted and enterprise tiers
LLM engineering platform for testing and evaluating AI agents in production
Tiered subscription with usage-based overage charges
Open-source AI orchestration framework for LLM applications
Open-source with optional paid enterprise support
The world's most popular data science platform
