Microsecond Inference &
Horizontal Scalability
All Inside Your Database
Get lightning speed feature recall for online predictions, along with instant scalability and long term monitoring for your machine learning models running inside the world’s most capable feature store.
How It
Works
PostgresML integrates the feature store, model store and inference layers into a seamless shared memory space.
Implemented inside PostgreSQL with Rust, it safely achieves maximum performance, eliminating network calls, process boundaries, data duplication and (de)serialization required in traditional machine learning architectures.
Try Now For Free
Fast
Up to 40x performance improvement over traditional machine learning architectures.
Comprehensive
PostgresML is your model store, feature store, inference server and deployment pipeline.
Simple
Say good bye to Python microservices. PostgresML trains and manages your machine learning models end to end.
Scalable
Scale your inference layer to millions of predictions per second with PostgresML replicas.
Your Data
Automatically add your features with replication from your primary data store, via our secure connection manager.
Highly Available
Abstract everything behind a single connection string with smart query routing, sharding, and enterprise-grade managed infrastructure.
How to Use It
Train
SELECT pgml.train(
'Sales Forecasting',
task => 'regression',
relation_name => 'historical_sales',
y_column_name => 'next_week_sales',
algorithm => 'xgboost'
);
Deploy
SELECT pgml.deploy(
'Sales Forecasting',
strategy => 'best_score',
algorithm => 'xgboost'
);
Predict
SELECT pgml.predict(
'Sales Forecasting',
ARRAY[
last_week_sales,
week_of_year
]
) AS prediction
ORDER BY prediction DESC;
What Adopters Say
"Absolutely brilliant" on PgCat, our sharded PostgreSQL proxy.
"The improvement is quite remarkable" on PostgresML v2.0.
"Bleeding edge stuff in a matter of minutes."
"I'm itching to apply it everywhere I set my foot!"
"The simplicity and ergonomics are really exciting."
Start Now
With Machine Learning
Get Started For Free
Discover All Features
Fastest Inference
PostgresML eliminates separation between your model server and your feature store, minimizing latency and computation costs. Our benchmarks show a 8x-40x improvement over Python HTTP microservices.
All Your Favorite Algorithms
Whether you need a simple linear regression or extreme gradient boosting, PostgresML includes support for all classification and regression algorithms in Scikit Learn, XGBoost, LightGBM and pre-trained deep learning models from Hugging Face.
Hyperparameter Search
Use either grid or random searches with cross validation on your training data to discover the most important knobs to tweak on your favorite algorithm, with best practices automatically enforced for testing model quality before deploying to production.
Online & Offline
Predictions are served via a standard Postgres connection to ensure that your core apps can always access both your data and your models in real time. Pure SQL workflows also enable batch predictions to cache results in native Postgres tables.
We Have the Perfect
Plan For You
Free
Free tier full of features for hobbyists.
Features
-
Open Source PostgresML with 47 machine learning algorithms
-
Automatic data preprocessing for building optimal models
-
5 GiB of disk space
-
Web Dashboard
-
PostgreSQL database to integrate with your apps
-
Hangout with us in Discord
Business
Empower your machine learning operations.
Features
-
Dedicated high performance hardware
-
Up to 16 TiB of disk space
-
Horizontally scalable inference with replicas
-
High availability for your production applications
-
Backups
-
Dedicated support
Enterprise
Tailored solutions for your enterprise.
Features
-
Everything in the Business Tier
-
Cloud Prem
-
Premium support and dedicated TAM
-
Influence on the roadmap
PostgresML
Powered by
PostgresML
